fix README section about python install

* README.md: structure and sort references
* bump v2.1.8

.github/workflows/pypi-deploy.yml

@@ -0,0 +1,101 @@
+
+name: CI Build Wheel
+
+on:
+  push:
+    # branches: [master]
+    tags: ["v[0-9]+.[0-9]+.[0-9]+"]
+
+jobs:
+
+  build_setup:
+    name: Prepare environment for wheel builds
+    runs-on: ubuntu-24.04
+    steps:
+      - uses: actions/checkout@v2
+      - name: Prepare wheel build
+        run: make -C python/ setup
+      - name: Store wheel configuration files
+        uses: actions/upload-artifact@v4.6.0
+        with:
+          name: wheel-config
+          path: python/
+      - name: Display files
+        run: ls -lR
+
+  build_wheels:
+    name: Build binary wheels on ${{ matrix.os }}
+    runs-on: ${{ matrix.os }}
+    needs: [build_setup]
+    strategy:
+      matrix:
+        os: [ubuntu-latest, windows-latest]
+
+    steps:
+      - uses: actions/checkout@v2
+      - uses: actions/setup-python@v2
+      - name: Install cibuildwheel
+        run: python -m pip install cibuildwheel==2.1.2
+      - name: Get wheel configuration files
+        uses: actions/download-artifact@v4.1.7
+        with:
+          name: wheel-config
+          path: python/
+      - name: Build wheels
+        run: python -m cibuildwheel --output-dir wheelhouse
+        working-directory: python/
+      - name: Store binary wheels
+        uses: actions/upload-artifact@v4.6.0
+        with:
+          name: binary-wheels-${{matrix.os}}-${{ strategy.job-index }}
+          path: python/wheelhouse/*.whl
+
+  build_sdist:
+    name: Build source distribution
+    runs-on: ubuntu-24.04
+    needs: [build_setup]
+    steps:
+      - uses: actions/checkout@v2
+      - name: Install cython
+        run: python -m pip install cython==0.29.24
+      - name: Get wheel configuration files
+        uses: actions/download-artifact@v4.1.7
+        with:
+          name: wheel-config
+          path: python/
+      - name: Build sdist
+        run: python setup.py sdist
+        working-directory: python/
+      - name: Store source wheels
+        uses: actions/upload-artifact@v4.6.0
+        with:
+          name: source-wheels
+          path: python/dist/*.tar.gz
+      - name: Display files
+        run: ls -lR
+
+
+  upload_pypi:
+    name: Upload wheels to PyPI
+    runs-on: ubuntu-24.04
+    needs: [build_wheels, build_sdist]
+
+    steps:
+      - name: Get source wheels
+        uses: actions/download-artifact@v4.1.7
+        with:
+          name: source-wheels
+          path: dist/
+      - name: Get binary wheels
+        uses: actions/download-artifact@v4.1.7
+        with:
+          path: dist/
+          pattern: binary-wheels-*
+          merge-multiple: true
+      - name: Display files
+        run: ls -lR
+
+      - uses: pypa/gh-action-pypi-publish@release/v1
+        with:
+          user: __token__
+          password: ${{ secrets.IMCTERMITE_GITHUB_WORKFLOW_PYPI_API_TOKEN }}

.gitignore

@@ -2,6 +2,7 @@
 eatraw
 eatdev
 imctermite
+IMCtermite
 
 nohup.out
 
@@ -15,6 +16,7 @@ cython/*.cpp
 
 *.log
 *.so
+*.pyd
 *.o
 *.csv
 *.parquet
@@ -29,3 +31,15 @@ pip/README.md
 pip/LICENSE
 pip/*egg-info
 pip/dist/
+pip/build/
+
+python/README.md
+python/README.rst
+python/LICENSE
+python/build
+python/*.egg-info
+python/dist
+python/*.soc
+python/lib/
+python/*.cpp
+python/wheelhouse/

README.md

@@ -1,14 +1,13 @@
 
-[![Total alerts](https://img.shields.io/lgtm/alerts/g/RecordEvolution/IMCtermite.svg?logo=lgtm&logoWidth=18)](https://lgtm.com/projects/g/RecordEvolution/IMCtermite/alerts/)
-[![Language grade: C/C++](https://img.shields.io/lgtm/grade/cpp/g/RecordEvolution/IMCtermite.svg?logo=lgtm&logoWidth=18)](https://lgtm.com/projects/g/RecordEvolution/IMCtermite/context:cpp)
-[![Language grade: Python](https://img.shields.io/lgtm/grade/python/g/RecordEvolution/IMCtermite.svg?logo=lgtm&logoWidth=18)](https://lgtm.com/projects/g/RecordEvolution/IMCtermite/context:python)
 [![LICENSE](https://img.shields.io/github/license/RecordEvolution/IMCtermite)](https://img.shields.io/github/license/RecordEvolution/IMCtermite)
 [![STARS](https://img.shields.io/github/stars/RecordEvolution/IMCtermite)](https://img.shields.io/github/stars/RecordEvolution/IMCtermite)
+![CI Build Wheel](https://github.com/RecordEvolution/IMCtermite/actions/workflows/pypi-deploy.yml/badge.svg?branch=&event=push)
+[![PYPI](https://img.shields.io/pypi/v/IMCtermite.svg)](https://pypi.org/project/imctermite/)
 
 # IMCtermite
 
 _IMCtermite_ provides access to the proprietary data format
-_IMC Bus Format_ with the file extension _.raw_ introduced and developed by
+_IMC2 Data Format_ with the file extension _.raw_ (or .dat) introduced and developed by
 [imc Test & Measurement GmbH](https://www.imc-tm.de/). This data format is
 employed i.a. by the measurement hardware
 [imc CRONOSflex](https://www.imc-tm.de/produkte/messtechnik-hardware/imc-cronosflex/ueberblick/)
@@ -19,7 +18,9 @@ for measurement data control and analysis. Thanks to the integrated Python modul
 the extracted measurement data can be stored in any open-source file format
 accessible by Python like i.a. _csv_, _json_ or _parquet_.
 
-On the [Record Evolution Platform](https://www.record-evolution.de/en/home-en/), the library can be used both as a command line tool for interactive usage and as a Python module to integrate the _.raw_  format into any ETL workflow.   
+On the [Record Evolution Platform](https://www.record-evolution.de/en/home-en/),
+the library can be used both as a command line tool for interactive usage and as a
+Python module to integrate the _.raw_  format into any ETL workflow.
 
 ## Overview
 
@@ -30,10 +31,11 @@ On the [Record Evolution Platform](https://www.record-evolution.de/en/home-en/),
 
 ## File format
 
-A data file of the _IMC Bus Format_ type with the extension _.raw_ is a _mixed text/binary
+A file of the _IMC2 Data Format_ type with extension _.raw_ (or .dat) is a _mixed text/binary
 file_ featuring a set of markers (keys) that indicate the start of various blocks
 of data that provide meta information and the actual measurement data. Every single
-marker is introduced by the character `"|" = 0x 7c` followed by two uppercase letters that characterize the type of marker. Each block is further divided into several
+marker is introduced by the character `"|" = 0x 7c` followed by two uppercase letters that
+characterize the type of marker. Each block is further divided into several
 parameters separated by commata `"," = 0x 2c` and terminated by a semicolon
 `";" = 0x 3b`. For instance, the header - first 600 bytes - of a raw file may
 look like this (in UTF-8 encoding):
@@ -129,28 +131,28 @@ which may require root permissions.
 
 ### Python
 
-To integrate the library into a customized ETL toolchain, several cython targets
+To integrate the library into a customized ETL toolchain, several python targets
 are available. For a local build that enables you to run the examples, use:
 
 ```
-make cython-build
+make python-build
 ```
 
-However, in a production environment, a proper installation of the module with
-`make cython-install` is recommended for system-wide availability of the module.
-
 #### Installation with pip
 
 The package is also available in the [Python Package Index](https://pypi.org)
-at [IMCtermite](https://pypi.org/project/IMCtermite/).
+at [imctermite](https://pypi.org/project/imctermite/).
 To install the latest version simply do
 
 ```Shell
-python3 -m pip install IMCtermite
+python3 -m pip install imctermite
 ```
 
-Note, that _python3_setuptools_ and _gcc version >= 10.2.0_ are required to
-successfully install and use it.
+which provides binary wheels for multiple architectures on _Windows_ and _Linux_
+and most _Python 3.x_ distributions. However, if your platform/architecture is
+not supported you can still compile the source distribution yourself, which
+requires _python3_setuptools_ and an up-to-date compiler supporting C++11
+standard (e.g. _gcc version >= 10.2.0_).
 
 ## Usage
 
@@ -186,23 +188,23 @@ options `imctermite sample-data.raw -b -c -s '|'`.
 
 ### Python
 
-Given the `imctermite` module is available, we can import it and declare an instance
+Given the `IMCtermite` module is available, we can import it and declare an instance
 of it by passing a _raw_ file to the constructor:
 
 ```Python
-import imc_termite
+import imctermite
 
-imcraw = imc_termite.imctermite(b"sample/sampleA.raw")
+imcraw = imctermite.imctermite(b"sample/sampleA.raw")
 ```
 
 An example of how to create an instance and obtain the list of channels is:
 
 ```Python
-import imc_termite
+import IMCtermite
 
 # declare and initialize instance of "imctermite" by passing a raw-file
 try :
-    imcraw = imc_termite.imctermite(b"samples/sampleA.raw")
+    imcraw = IMCtermite.imctermite(b"samples/sampleA.raw")
 except RuntimeError as e :
     print("failed to load/parse raw-file: " + str(e))
 
@@ -211,15 +213,37 @@ channels = imcraw.get_channels(False)
 print(channels)
 ```
 
-A more complete [example](python/usage.py), including the methods for obtaining the
-channels, i.a. their data and/or directly printing them to files, can be found
-in the Python folder.
+A more complete [example](python/examples/usage.py), including the methods for
+obtaining the channels, i.a. their data and/or directly printing them to files,
+can be found in the `python/examples` folder.
 
 ## References
 
+### IMC
+
 - https://www.imc-tm.de/produkte/messtechnik-software/imc-famos/funktionen/im-und-export/
 - https://www.imc-tm.de/produkte/messtechnik-hardware/imc-cronosflex/ueberblick/
-- https://cython.readthedocs.io/en/latest/src/userguide/wrapping_CPlusPlus.html
+- https://www.imc-tm.de/download-center/produkt-downloads/imc-famos/handbuecher
+- https://www.imc-tm.de/fileadmin/Public/Downloads/Manuals/imc_FAMOS/imcGemeinsameKomponenten.pdf
 - https://github.com/Apollo3zehn/ImcFamosFile
 - https://apollo3zehn.github.io/ImcFamosFile/api/ImcFamosFile.FamosFileKeyType.html
+
+### Cython
+
+- https://cython.readthedocs.io/en/latest/src/userguide/wrapping_CPlusPlus.html
+
+### PyPI
+
 - https://pypi.org/help/#apitoken
+- https://sgoel.dev/posts/uploading-binary-wheels-to-pypi-from-github-actions/
+- https://docs.github.com/en/actions/reference/workflow-syntax-for-github-actions#jobsjob_idstepsrun
+- https://github.com/pypa/cibuildwheel/blob/main/examples/github-deploy.yml
+- https://cibuildwheel.readthedocs.io/en/stable/deliver-to-pypi/
+- https://github.com/actions/download-artifact#download-all-artifacts
+- https://github.com/actions/download-artifact?tab=readme-ov-file#download-multiple-filtered-artifacts-to-the-same-directory
+
+### iconv
+
+- https://www.gnu.org/software/libiconv/
+- https://vcpkg.io/en/packages.html
+- https://vcpkg.io/en/getting-started

assets/imctermite.svg

@@ -1,111 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<svg
-   xmlns:dc="http://purl.org/dc/elements/1.1/"
-   xmlns:cc="http://creativecommons.org/ns#"
-   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
-   xmlns:svg="http://www.w3.org/2000/svg"
-   xmlns="http://www.w3.org/2000/svg"
-   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
-   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
-   height="78.080002"
-   width="325.72"
-   id="svg3945"
-   version="1.1"
-   viewBox="0 0 325.72 78.08"
-   sodipodi:docname="imctermite.svg"
-   inkscape:version="1.0.1 (3bc2e813f5, 2020-09-07)">
-  <sodipodi:namedview
-     pagecolor="#ffffff"
-     bordercolor="#666666"
-     borderopacity="1"
-     objecttolerance="10"
-     gridtolerance="10"
-     guidetolerance="10"
-     inkscape:pageopacity="0"
-     inkscape:pageshadow="2"
-     inkscape:window-width="2048"
-     inkscape:window-height="1088"
-     id="namedview18"
-     showgrid="false"
-     inkscape:zoom="2.5577614"
-     inkscape:cx="138.13369"
-     inkscape:cy="-27.086877"
-     inkscape:window-x="0"
-     inkscape:window-y="32"
-     inkscape:window-maximized="1"
-     inkscape:current-layer="svg3945"
-     inkscape:document-rotation="0" />
-  <metadata
-     id="metadata3951">
-    <rdf:RDF>
-      <cc:Work
-         rdf:about="">
-        <dc:format>image/svg+xml</dc:format>
-        <dc:type
-           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
-        <dc:title>flasher</dc:title>
-      </cc:Work>
-    </rdf:RDF>
-  </metadata>
-  <defs
-     id="defs3949" />
-  <title
-     id="title3916">flasher</title>
-  <g
-     id="logog"
-     transform="translate(0,0.99981694)">
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="m 32.86,2 -13,7.5 v 0 h -0.05 v 0 l -0.48,0.28 c -4.27,2.46 -5.68,11.38 -6.06,14.75 L 36.2,11.33 c 0.39,-0.19 7.6,-3.69 13.57,-3.69 h 0.14 L 40.13,2 a 8.15,8.15 0 0 0 -7.27,0"
-       id="path138"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="M 5.68,17.69 A 8.2,8.2 0 0 0 2,24 v 15.78 c 0,4.9 7,10.48 9.75,12.46 V 25.77 c 0,-0.44 0.6,-8.55 3.65,-13.72 z"
-       id="path142"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="m 12.1,54.12 v 0 C 11.74,53.88 5,49.41 2,44.24 v 11.14 a 8.2,8.2 0 0 0 3.64,6.3 l 13.5,7.79 c 4.28,2.46 12.7,-0.77 15.81,-2.12 z"
-       id="path146"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="m 36.79,68 c -0.4,0.19 -7.71,3.75 -13.71,3.69 l 9.78,5.64 a 8.15,8.15 0 0 0 7.27,0 l 13.51,-7.8 c 4.27,-2.46 5.68,-11.39 6.06,-14.75 z"
-       id="path150"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="M 61.2,27.13 V 53.6 c 0,0.44 -0.6,8.55 -3.65,13.72 l 9.77,-5.64 A 8.2,8.2 0 0 0 71,55.38 V 39.59 c 0,-4.94 -7,-10.5 -9.75,-12.46"
-       id="path154"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="M 67.31,17.69 53.81,9.9 C 49.53,7.44 41.11,10.67 38,12 l 22.85,13.23 v 0 a 43.43,43.43 0 0 1 5.7,4.51 24,24 0 0 1 4.45,5.35 V 24 a 8.2,8.2 0 0 0 -3.64,-6.3"
-       id="path158"
-       inkscape:connector-curvature="0" />
-  </g>
-  <g
-     id="re"
-     transform="translate(0,0.99981694)" />
-  <text
-     id="text3955"
-     y="55.47554"
-     x="74.101189"
-     style="font-style:normal;font-weight:normal;font-size:40px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#364d5c;fill-opacity:1;stroke:none"
-     xml:space="preserve"><tspan
-       style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:44px;font-family:sans-serif;-inkscape-font-specification:'sans-serif, Bold';font-variant-ligatures:normal;font-variant-caps:normal;font-variant-numeric:normal;font-feature-settings:normal;text-align:start;writing-mode:lr-tb;text-anchor:start;fill:#364d5c;fill-opacity:1"
-       y="55.47554"
-       x="74.101189"
-       id="tspan3953"><tspan
-   id="tspan24"
-   style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold'">IMC</tspan><tspan
-   id="tspan3845"
-   style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:sans-serif;-inkscape-font-specification:sans-serif">termite</tspan> </tspan></text>
-</svg>

assets/raweater.svg

@@ -1,112 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<svg
-   xmlns:dc="http://purl.org/dc/elements/1.1/"
-   xmlns:cc="http://creativecommons.org/ns#"
-   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
-   xmlns:svg="http://www.w3.org/2000/svg"
-   xmlns="http://www.w3.org/2000/svg"
-   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
-   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
-   height="78.080002"
-   width="290.72"
-   id="svg3945"
-   version="1.1"
-   viewBox="0 0 290.72 78.08"
-   sodipodi:docname="raweater.svg"
-   inkscape:version="0.92.5 (2060ec1f9f, 2020-04-08)">
-  <sodipodi:namedview
-     pagecolor="#ffffff"
-     bordercolor="#666666"
-     borderopacity="1"
-     objecttolerance="10"
-     gridtolerance="10"
-     guidetolerance="10"
-     inkscape:pageopacity="0"
-     inkscape:pageshadow="2"
-     inkscape:window-width="1360"
-     inkscape:window-height="704"
-     id="namedview18"
-     showgrid="false"
-     inkscape:zoom="0.90430522"
-     inkscape:cx="191.86"
-     inkscape:cy="38.540001"
-     inkscape:window-x="0"
-     inkscape:window-y="27"
-     inkscape:window-maximized="1"
-     inkscape:current-layer="svg3945" />
-  <metadata
-     id="metadata3951">
-    <rdf:RDF>
-      <cc:Work
-         rdf:about="">
-        <dc:format>image/svg+xml</dc:format>
-        <dc:type
-           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
-        <dc:title>flasher</dc:title>
-      </cc:Work>
-    </rdf:RDF>
-  </metadata>
-  <defs
-     id="defs3949" />
-  <title
-     id="title3916">flasher</title>
-  <g
-     id="logog"
-     transform="translate(0,0.99981694)">
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="m 32.86,2 -13,7.5 v 0 h -0.05 v 0 l -0.48,0.28 c -4.27,2.46 -5.68,11.38 -6.06,14.75 L 36.2,11.33 c 0.39,-0.19 7.6,-3.69 13.57,-3.69 h 0.14 L 40.13,2 a 8.15,8.15 0 0 0 -7.27,0"
-       id="path138"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="M 5.68,17.69 A 8.2,8.2 0 0 0 2,24 v 15.78 c 0,4.9 7,10.48 9.75,12.46 V 25.77 c 0,-0.44 0.6,-8.55 3.65,-13.72 z"
-       id="path142"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="m 12.1,54.12 v 0 C 11.74,53.88 5,49.41 2,44.24 v 11.14 a 8.2,8.2 0 0 0 3.64,6.3 l 13.5,7.79 c 4.28,2.46 12.7,-0.77 15.81,-2.12 z"
-       id="path146"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="m 36.79,68 c -0.4,0.19 -7.71,3.75 -13.71,3.69 l 9.78,5.64 a 8.15,8.15 0 0 0 7.27,0 l 13.51,-7.8 c 4.27,-2.46 5.68,-11.39 6.06,-14.75 z"
-       id="path150"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="M 61.2,27.13 V 53.6 c 0,0.44 -0.6,8.55 -3.65,13.72 l 9.77,-5.64 A 8.2,8.2 0 0 0 71,55.38 V 39.59 c 0,-4.94 -7,-10.5 -9.75,-12.46"
-       id="path154"
-       inkscape:connector-curvature="0" />
-    <path
-       style="fill:#364d5c"
-       transform="translate(-2.04,-1.15)"
-       d="M 67.31,17.69 53.81,9.9 C 49.53,7.44 41.11,10.67 38,12 l 22.85,13.23 v 0 a 43.43,43.43 0 0 1 5.7,4.51 24,24 0 0 1 4.45,5.35 V 24 a 8.2,8.2 0 0 0 -3.64,-6.3"
-       id="path158"
-       inkscape:connector-curvature="0" />
-  </g>
-  <g
-     id="re"
-     transform="translate(0,0.99981694)" />
-  <text
-     id="text3955"
-     y="55.47554"
-     x="74.101189"
-     style="font-style:normal;font-weight:normal;font-size:40px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#364d5c;fill-opacity:1;stroke:none"
-     xml:space="preserve"><tspan
-       style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:44px;font-family:sans-serif;-inkscape-font-specification:'sans-serif, Bold';font-variant-ligatures:normal;font-variant-caps:normal;font-variant-numeric:normal;font-feature-settings:normal;text-align:start;writing-mode:lr-tb;text-anchor:start;fill:#364d5c;fill-opacity:1"
-       y="55.47554"
-       x="74.101189"
-       id="tspan3953"><tspan
-   id="tspan24"
-   style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold'">R</tspan><tspan
-   id="tspan3845"
-   style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-family:sans-serif;-inkscape-font-specification:sans-serif">aw<tspan
-   style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold'"
-   id="tspan22">E</tspan>ater</tspan> </tspan></text>
-</svg>

cython/raw_eater.pxd

@@ -1,41 +0,0 @@
-# cython: language_level = 3
-# distutils: language = c++
-
-# use some C++ STL libraries
-from libcpp.string cimport string
-from libcpp.vector cimport vector
-from libcpp cimport bool
-
-# to include implemenation/definition file
-#cdef extern from "raweat.cpp":
-#    pass
-
-# these method names have to match the C definitions of the methods!!
-#
-# for how to overload the constructor see
-# https://cython.readthedocs.io/en/latest/src/userguide/wrapping_CPlusPlus.html
-# and propagating exceptions from C++ to Python
-# http://docs.cython.org/en/latest/src/userguide/wrapping_CPlusPlus.html#exceptions
-
-cdef extern from "../lib/raweat.hpp":
-    cdef cppclass raw_eater:
-        # constructor(s)
-        raw_eater() except +
-        raw_eater(string) except +
-        # set new file for decoding
-        void set_file(string)
-        # perform conversion (pass any C++ exceptions to Python)
-        void setup_and_conversion() except +
-        # get validity of data format
-        bool get_valid()
-        # get channel name and unit
-        string get_name()
-        string get_unit()
-        # get time step and time unit
-        double get_dt()
-        string get_temp_unit()
-        # get data array of time and measured quantity's channel
-        vector[double] get_time()
-        vector[double] get_data()
-        # dump all data to .csv
-        void write_table(const char*,char delimiter)

cython/raw_eater.pyx

@@ -1,58 +0,0 @@
-
-from raw_eater cimport raweater
-import numpy as np
-import re
-import os
-
-cdef class raweater:
-
-  # C++ instance of class => stack allocated (requires nullary constructor!)
-  cdef raw_eater rawit
-
-  # pointer to C++ instance (if there's no nullary constructor)
-  # cdef raw_eater *rawit
-
-  def __cinit__(self, string rawfile = b''):
-    if rawfile.decode() == "":
-      self.rawit = raw_eater()
-      # self.rawit = new raw_eater()
-    else:
-      if not os.path.isfile(rawfile) :
-        raise ValueError("'" + str(rawfile) + "' does not exist")
-      self.rawit = raw_eater(rawfile)
-      # self.rawit = new raw_eater(rawfile)
-
-  # def __dealloc__(self):
-  #   del self.rawit
-
-  def set_file(self, string rawfile):
-    if not os.path.isfile(rawfile) :
-      raise ValueError("'" + str(rawfile) + "' does not exist")
-    self.rawit.set_file(rawfile)
-
-  def do_conversion(self):
-      self.rawit.setup_and_conversion()
-
-  def validity(self):
-    return self.rawit.get_valid()
-
-  def channel_name(self):
-    return self.rawit.get_name()
-
-  def unit(self):
-    return self.rawit.get_unit()
-
-  def dt(self):
-    return self.rawit.get_dt()
-
-  def time_unit(self):
-    return self.rawit.get_temp_unit()
-
-  def get_time(self):
-    return self.rawit.get_time()
-
-  def get_channel(self):
-    return self.rawit.get_data()
-
-  def write_table(self, const char* csvfile, char delimiter):
-    self.rawit.write_table(csvfile,delimiter)

cython/raw_meat.pxd

@@ -1,37 +0,0 @@
-# cython: language_level = 3
-# distutils: language = c++
-
-# use some C++ STL libraries
-from libcpp.string cimport string
-from libcpp.vector cimport vector
-from libcpp cimport bool
-
-# these method names have to match the C++ definitions of the methods!!
-cdef extern from "../lib/rawmerge.hpp":
-    cdef cppclass raw_merger:
-        raw_merger(string) except +
-        # get validity of data format
-        bool get_valid()
-        # get channel name and unit
-        string get_name()
-        string get_unit()
-        # get time step and time unit
-        double get_dt()
-        string get_temp_unit()
-        # get data array of time and measured quantity's channel
-        vector[double] get_time()
-        vector[double] get_data()
-        # dump all data to .csv
-        void write_table(const char*,char)
-        # add channel and try to merge it (pass C++ exceptions to Python)
-        bool add_channel(string) except +
-        # get total number of (added) channels
-        int get_num_channels()
-        # get list of channel names
-        vector[string] get_channel_names()
-        # get data of particular channel
-        vector[double] get_channel(int)
-        # get total merged time series
-        vector[double] get_time_series()
-        # dump all channels to .csv
-        void write_table_all(const char*,char)

cython/raw_meat.pyx

@@ -1,58 +0,0 @@
-
-# from <raw_meat> has to match name of .pxd file and cimport name of class defined in .pxd
-from raw_meat cimport raw_merger
-import numpy as np
-import re
-
-cdef class rawmerger:
-
-    # pointer to C++ instance (since there's no nullary constructor)
-    cdef raw_merger *rawit
-
-    def __cinit__(self, string rawfile):
-        self.rawit = new raw_merger(rawfile)
-
-    def __dealloc__(self):
-        del self.rawit
-
-    def validity(self):
-        return self.rawit.get_valid()
-
-    def channel_name(self):
-        return self.rawit.get_name()
-
-    def unit(self):
-        return self.rawit.get_unit()
-
-    def dt(self):
-        return self.rawit.get_dt()
-
-    def time_unit(self):
-        return self.rawit.get_temp_unit()
-
-    def get_time(self):
-        return self.rawit.get_time()
-
-    def get_channel(self):
-        return self.rawit.get_data()
-
-    def write_table(self, const char* csvfile, char delimiter):
-        return self.rawit.write_table(csvfile,delimiter)
-
-    def add_channel(self, string rawfile):
-        return self.rawit.add_channel(rawfile)
-
-    def get_num_channels(self):
-        return self.rawit.get_num_channels()
-
-    def get_channel_names(self):
-        return self.rawit.get_channel_names()
-
-    def get_channel_by_index(self, int chidx):
-        return self.rawit.get_channel(chidx)
-
-    def get_time_series(self):
-        return self.rawit.get_time_series()
-
-    def write_table_all(self, const char* csvfile, char delimiter):
-        return self.rawit.write_table_all(csvfile,delimiter)

cython/setup.py

@@ -1,24 +0,0 @@
-from distutils.core import setup
-from distutils.extension import Extension
-from Cython.Build import cythonize
-
-extensions = Extension(
-    name="imc_termite",
-    sources=["cython/py_imc_termite.pyx"],
-    # libraries=[""],
-    # library_dirs=["lib"],
-    include_dirs=["lib"],
-    language='c++',
-    extra_compile_args=['-std=c++17','-Wno-unused-variable'],
-    extra_link_args=['-std=c++17'],
-)
-
-setup(
-    name="imc_termite",
-    version='1.2.5',
-    description='IMCtermite cython extension',
-    author='Record Evolution GmbH',
-    author_email='mario.fink@record-evolution.de',
-    url='https://github.com/RecordEvolution/IMCtermite.git',
-    ext_modules=cythonize(extensions,force=True)
-)

cython/setup_raw_eater.py

@@ -1,20 +0,0 @@
-from distutils.core import setup
-from distutils.extension import Extension
-from Cython.Build import cythonize
-
-extensions = Extension(
-    name="raw_eater",
-    sources=["cython/raw_eater.pyx"],
-    # libraries=[""],
-    library_dirs=["src"],
-    include_dirs=["src"],
-    language='c++',
-    extra_compile_args=['-std=c++11','-Wno-unused-variable'],
-    extra_link_args=['-std=c++11'],
-    #extra_objects=["lib/parquet/libarrow.so.200.0.0"],
-)
-
-setup(
-    name="raw_eater",
-    ext_modules=cythonize(extensions)
-)

cython/setup_raw_meat.py

@@ -1,20 +0,0 @@
-from distutils.core import setup
-from distutils.extension import Extension
-from Cython.Build import cythonize
-
-extensions = Extension(
-    name="raw_meat",
-    sources=["cython/raw_meat.pyx"],
-    # libraries=[""],
-    library_dirs=["src"],
-    include_dirs=["src"],
-    language='c++',
-    extra_compile_args=['-std=c++11','-Wno-unused-variable'],
-    extra_link_args=['-std=c++11'],
-    #extra_objects=["lib/parquet/libarrow.so.200.0.0"],
-)
-
-setup(
-    name="raw_meat",
-    ext_modules=cythonize(extensions)
-)

lib/imc_block.hpp

@@ -63,7 +63,7 @@ namespace imc
         std::cout<<"WARNING: invalid length parameter in "<<thekey_.name_<<"-block "
                  <<"(block-end:"<<end_<<",buffer-size:"<<buffer_->size()<<")"
                  <<" => resetting block-end to buffer-size\n";
-        end_ = buffer_->size();
+        end_ = (unsigned long int)(buffer_->size());
       }
 
       try {

lib/imc_channel.hpp

@@ -3,41 +3,80 @@
 #ifndef IMCCHANNEL
 #define IMCCHANNEL
 
-#include <sstream>
 #include "imc_datatype.hpp"
 #include "imc_conversion.hpp"
+#include "imc_block.hpp"
+#include <sstream>
+#include <math.h>
+#include <chrono>
+#include <ctime>
+#include <time.h>
+#if defined(__linux__) || defined(__APPLE__)
+#include <iconv.h>
+#elif defined(__WIN32__) || defined(_WIN32)
+#define timegm _mkgmtime
+#endif
 
 //---------------------------------------------------------------------------//
 
 namespace imc
 {
+  struct component_env
+  {
+    std::string uuid_;
+    // required channel components for CG channels only
+    std::string CCuuid_, CPuuid_;
+    // optional channel components for CG channels only
+    std::string CDuuid_, NTuuid_;
+    std::string Cbuuid_, CRuuid_;
+
+    // reset all members
+    void reset()
+    {
+      uuid_.clear();
+      CCuuid_.clear();
+      CPuuid_.clear();
+      CDuuid_.clear();
+      Cbuuid_.clear();
+      CRuuid_.clear();
+      NTuuid_.clear();
+    }
+  };
+
   // collect uuid's of blocks required for full channel reconstruction
   struct channel_env
   {
     // define unique identifer for channel_env
     std::string uuid_;
 
+    // collect common affiliate blocks for every channel
+    std::string NOuuid_, NLuuid_;
     // collect affiliate blocks for a single channel
-    std::string CBuuid_, CGuuid_, CCuuid_, CNuuid_;
-    std::string CDuuid_, CTuuid_, Cbuuid_, CPuuid_, CRuuid_, CSuuid_;
-    std::string NTuuid_, NOuuid_;
+    // channel types
+    std::string CBuuid_, CGuuid_, CIuuid_, CTuuid_;
+    std::string CNuuid_, CDuuid_, NTuuid_;
+    std::string CSuuid_;
+
+    component_env compenv1_;
+    component_env compenv2_;
+
 
     // reset all members
     void reset()
     {
       uuid_.clear();
+      NOuuid_.clear();
+      NLuuid_.clear();
       CBuuid_.clear();
       CGuuid_.clear();
-      CCuuid_.clear();
+      CIuuid_.clear();
+      CTuuid_.clear();
       CNuuid_.clear();
       CDuuid_.clear();
-      CTuuid_.clear();
-      Cbuuid_.clear();
-      CPuuid_.clear();
-      CRuuid_.clear();
-      CSuuid_.clear();
       NTuuid_.clear();
-      NOuuid_.clear();
+      CSuuid_.clear();
+      compenv1_.reset();
+      compenv2_.reset();
     }
 
     // get info
@@ -45,20 +84,23 @@ namespace imc
     {
       std::stringstream ss;
       ss<<std::setw(width)<<std::left<<"uuid:"<<uuid_<<"\n"
+        <<std::setw(width)<<std::left<<"NOuuid:"<<NOuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"NLuuid:"<<NLuuid_<<"\n"
+        //
         <<std::setw(width)<<std::left<<"CBuuid:"<<CBuuid_<<"\n"
         <<std::setw(width)<<std::left<<"CGuuid:"<<CGuuid_<<"\n"
-        <<std::setw(width)<<std::left<<"CCuuid:"<<CCuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"CIuuid:"<<CIuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"CTuuid:"<<CTuuid_<<"\n"
         <<std::setw(width)<<std::left<<"CNuuid:"<<CNuuid_<<"\n"
         //
-        <<std::setw(width)<<std::left<<"CDuuid:"<<CDuuid_<<"\n"
-        <<std::setw(width)<<std::left<<"CTuuid:"<<CTuuid_<<"\n"
-        <<std::setw(width)<<std::left<<"Cbuuid:"<<Cbuuid_<<"\n"
-        <<std::setw(width)<<std::left<<"CPuuid:"<<CPuuid_<<"\n"
-        <<std::setw(width)<<std::left<<"CRuuid:"<<CRuuid_<<"\n"
-        <<std::setw(width)<<std::left<<"CSuuid:"<<CSuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"CCuuid:"<<compenv1_.CCuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"CPuuid:"<<compenv1_.CPuuid_<<"\n"
         //
-        <<std::setw(width)<<std::left<<"NTuuid:"<<NTuuid_<<"\n"
-        <<std::setw(width)<<std::left<<"NOuuid:"<<NOuuid_<<"\n";
+        <<std::setw(width)<<std::left<<"CDuuid:"<<compenv1_.CDuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"Cbuuid:"<<compenv1_.Cbuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"CRuuid:"<<compenv1_.CRuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"NTuuid:"<<compenv1_.NTuuid_<<"\n"
+        <<std::setw(width)<<std::left<<"CSuuid:"<<CSuuid_<<"\n";
       return ss.str();
     }
 
@@ -67,44 +109,69 @@ namespace imc
     {
       std::stringstream ss;
       ss<<"{"<<"\"uuid\":\""<<uuid_
+             <<"\",\"NOuuid\":\""<<NOuuid_
+             <<"\",\"NLuuid\":\""<<NLuuid_
              <<"\",\"CBuuid\":\""<<CBuuid_
              <<"\",\"CGuuid\":\""<<CGuuid_
-             <<"\",\"CCuuid\":\""<<CCuuid_
-             <<"\",\"CNuuid\":\""<<CNuuid_
-             <<"\",\"CDuuid\":\""<<CDuuid_
+             <<"\",\"CIuuid\":\""<<CIuuid_
              <<"\",\"CTuuid\":\""<<CTuuid_
-             <<"\",\"Cbuuid\":\""<<Cbuuid_
-             <<"\",\"CPuuid\":\""<<CPuuid_
-             <<"\",\"CRuuid\":\""<<CRuuid_
+             <<"\",\"CNuuid\":\""<<CNuuid_
+             <<"\",\"CCuuid\":\""<<compenv1_.CCuuid_
+             <<"\",\"CPuuid\":\""<<compenv1_.CPuuid_
+             <<"\",\"CDuuid\":\""<<compenv1_.CDuuid_
+             <<"\",\"Cbuuid\":\""<<compenv1_.Cbuuid_
+             <<"\",\"CRuuid\":\""<<compenv1_.CRuuid_
+             <<"\",\"NTuuid\":\""<<compenv1_.NTuuid_
              <<"\",\"CSuuid\":\""<<CSuuid_
-             <<"\",\"NTuuid\":\""<<NTuuid_
-             <<"\",\"NOuuid\":\""<<NOuuid_
              <<"\"}";
       return ss.str();
     }
   };
 
+  // adjust stream object
+  void customize_stream(std::ostream& stout, int prec, bool fixed)
+  {
+    if ( fixed )
+    {
+      stout<<std::setprecision(prec)<<std::fixed;
+    }
+    else
+    {
+      stout<<std::setprecision(prec);
+    }
+  }
+
   // given a list of numeric objects, join it into a string
   template<typename dt>
-  std::string joinvec(std::vector<dt> myvec, unsigned long int limit = 10)
+  std::string joinvec(std::vector<dt> myvec, unsigned long int limit = 10, int prec = 10, bool fixed = true)
   {
-    // include entire list for limit = - 1
-    limit = (limit == 0) ? myvec.size() : limit;
+    // include entire list for limit = 0
+    unsigned long int myvecsize = (unsigned long int)myvec.size();
+    limit = (limit == 0) ? myvecsize : limit;
 
     std::stringstream ss;
     ss<<"[";
     if ( myvec.size() <= limit )
     {
-      for ( dt el: myvec ) ss<<std::setprecision(10)<<el<<",";
+      for ( dt el: myvec )
+      {
+        customize_stream(ss,prec,fixed);
+        ss<<el<<",";
+      }
     }
     else
     {
-      unsigned long int heals = (unsigned long int)(limit/2.);
-      for ( unsigned long int i = 0; i < heals; i++ ) ss<<myvec[i]<<",";
-      ss<<"...";
-      for ( unsigned long int i = myvec.size()-heals; i < myvec.size(); i++ )
+      unsigned long int heals = limit/2;
+      for ( unsigned long int i = 0; i < heals; i++ )
       {
-        ss<<std::setprecision(10)<<myvec[i]<<",";
+        customize_stream(ss,prec,fixed);
+        ss<<myvec[i]<<",";
+      }
+      ss<<"...";
+      for ( unsigned long int i = myvecsize-heals; i < myvecsize; i++ )
+      {
+        customize_stream(ss,prec,fixed);
+        ss<<myvec[i]<<",";
       }
     }
     std::string sumstr = ss.str();
@@ -113,6 +180,131 @@ namespace imc
     return sumstr;
   }
 
+  #if defined(__linux__) || defined(__APPLE__)
+  // convert encoding of any descriptions, channel-names, units etc.
+  class iconverter
+  {
+    std::string in_enc_, out_enc_;
+    iconv_t cd_;
+    size_t out_buffer_size_;
+
+    public:
+
+      iconverter(std::string in_enc, std::string out_enc, size_t out_buffer_size = 1024) :
+        in_enc_(in_enc), out_enc_(out_enc), out_buffer_size_(out_buffer_size)
+      {
+        // allocate descriptor for character set conversion
+        // (https://man7.org/linux/man-pages/man3/iconv_open.3.html)
+        cd_ = iconv_open(out_enc.c_str(), in_enc.c_str());
+
+        if ( (iconv_t)-1 == cd_ )
+        {
+          if ( errno == EINVAL )
+          {
+            std::string errmsg = std::string("The encoding conversion from ") + in_enc
+              + std::string(" to ") + out_enc + std::string(" is not supported by the implementation.");
+            throw std::runtime_error(errmsg);
+          }
+        }
+      }
+
+      void convert(std::string &astring)
+      {
+        if ( astring.empty() ) return;
+
+        std::vector<char> in_buffer(astring.begin(),astring.end());
+        char *inbuf = &in_buffer[0];
+        size_t inbytes = in_buffer.size();
+
+        std::vector<char> out_buffer(out_buffer_size_);
+        char *outbuf = &out_buffer[0];
+        size_t outbytes = out_buffer.size();
+
+        // perform character set conversion
+        // ( - https://man7.org/linux/man-pages/man3/iconv.3.html
+        //   - https://www.ibm.com/docs/en/zos/2.2.0?topic=functions-iconv-code-conversion )
+        while ( inbytes > 0 )
+        {
+          size_t res = iconv(cd_,&inbuf,&inbytes,&outbuf,&outbytes);
+
+          if ( (size_t)-1 == res )
+          {
+            std::string errmsg;
+            if ( errno == EILSEQ )
+            {
+              errmsg = std::string("An invalid multibyte sequence is encountered in the input.");
+              throw std::runtime_error(errmsg);
+            }
+            else if ( errno == EINVAL )
+            {
+              errmsg = std::string("An incomplete multibyte sequence is encountered in the input")
+                     + std::string(" and the input byte sequence terminates after it.");
+            }
+            else if ( errno == E2BIG )
+            {
+              errmsg = std::string("The output buffer has no more room for the next converted character.");
+            }
+            throw std::runtime_error(errmsg);
+          }
+        }
+
+        std::string outstring(out_buffer.begin(),out_buffer.end()-outbytes);
+        astring = outstring;
+      }
+  };
+  #elif defined(__WIN32__) || defined(_WIN32)
+  class iconverter
+  {
+    public:
+      iconverter(std::string in_enc, std::string out_enc, size_t out_buffer_size = 1024) {}
+      void convert(std::string &astring) {}
+  };
+  #endif
+
+  struct component_group
+  {
+    imc::component CC_;
+    imc::packaging CP_;
+    imc::abscissa CD_;
+    imc::buffer Cb_;
+    imc::range CR_;
+    imc::channelobj CN_;
+    imc::triggertime NT_;
+
+    component_env compenv_;
+
+    // Constructor to parse the associated blocks
+    component_group(component_env &compenv, std::map<std::string, imc::block>* blocks, std::vector<unsigned char>* buffer)
+        : compenv_(compenv)
+    {
+        if (blocks->count(compenv.CCuuid_) == 1)
+        {
+            CC_.parse(buffer, blocks->at(compenv.CCuuid_).get_parameters());
+        }
+        if (blocks->count(compenv.CPuuid_) == 1)
+        {
+            CP_.parse(buffer, blocks->at(compenv.CPuuid_).get_parameters());
+        }
+        if (blocks->count(compenv.CDuuid_) == 1)
+        {
+            CD_.parse(buffer, blocks->at(compenv.CDuuid_).get_parameters());
+        }
+        if (blocks->count(compenv.Cbuuid_) == 1)
+        {
+            Cb_.parse(buffer, blocks->at(compenv.Cbuuid_).get_parameters());
+        }
+        if (blocks->count(compenv.CRuuid_) == 1)
+        {
+            CR_.parse(buffer, blocks->at(compenv.CRuuid_).get_parameters());
+        }
+        if (blocks->count(compenv.NTuuid_) == 1)
+        {
+            NT_.parse(buffer, blocks->at(compenv.NTuuid_).get_parameters());
+        }
+    }
+  };
+
+
   // channel
   struct channel
   {
@@ -121,231 +313,385 @@ namespace imc
     std::map<std::string,imc::block>* blocks_;
     std::vector<unsigned char>* buffer_;
 
+    imc::origin_data NO_;
+    imc::language NL_;
+    imc::text CT_;
+    imc::groupobj CB_;
+    imc::datafield CG_;
+    imc::channelobj CN_;
+
     // collect meta-data of channels according to env,
-    //  just everything valueable in here
+    // just everything valueable in here
+    // TODO: is this necessary?
     std::string uuid_;
     std::string name_, comment_;
     std::string origin_, origin_comment_, text_;
+    std::chrono::system_clock::time_point trigger_time_, absolute_trigger_time_;
+    double trigger_time_frac_secs_;
+    std::string language_code_, codepage_;
     std::string yname_, yunit_;
     std::string xname_, xunit_;
-    double xstepwidth_, xoffset_;
+    double xstepwidth_, xstart_;
+    int xprec_;
+    int dimension_;
 
     // buffer and data
-    int signbits_, num_bytes_;
+    int xsignbits_, xnum_bytes_;
+    int ysignbits_, ynum_bytes_;
     // unsigned long int byte_offset_;
-    unsigned long int buffer_offset_, buffer_size_;
-    int datatp_;
-    imc::datatype dattyp_;
-    std::vector<imc::datatype> ydata_;
-    std::vector<double> xdata_;
+    unsigned long int xbuffer_offset_, ybuffer_offset_;
+    unsigned long int xbuffer_size_, ybuffer_size_;
+    long int addtime_;
+    imc::numtype xdatatp_, ydatatp_;
+    std::vector<imc::datatype> xdata_, ydata_;
 
     // range, factor and offset
-    double factor_, offset_;
+    double xfactor_, yfactor_;
+    double xoffset_, yoffset_;
 
     // group reference the channel belongs to
-    int group_index_;
+    unsigned long int group_index_;
     std::string group_uuid_, group_name_, group_comment_;
 
     // constructor takes channel's block environment
     channel(channel_env &chnenv, std::map<std::string,imc::block>* blocks,
                                  std::vector<unsigned char>* buffer):
       chnenv_(chnenv), blocks_(blocks), buffer_(buffer),
-      factor_(1.), offset_(0.),
+      xfactor_(1.), yfactor_(1.), xoffset_(0.), yoffset_(0.),
       group_index_(-1)
     {
-      // declare list of block parameters
-      std::vector<imc::parameter> prms;
-
       // use uuid from CN block
       uuid_ = chnenv_.CNuuid_;
 
+      // extract associated NO data
+      if ( blocks_->count(chnenv_.NOuuid_) == 1 )
+      {
+	NO_.parse(buffer_, blocks_->at(chnenv_.NOuuid_).get_parameters());
+	origin_ = NO_.generator_;
+	comment_ = NO_.comment_;
+      }
+
+      // extract associated NL data
+      if ( blocks_->count(chnenv_.NLuuid_) == 1 )
+      {
+	NL_.parse(buffer_, blocks_->at(chnenv_.NLuuid_).get_parameters());
+	codepage_ = NL_.codepage_;
+	language_code_ = NL_.language_code_;
+      }
+
       // extract associated CB data
       if ( blocks_->count(chnenv_.CBuuid_) == 1 )
       {
-        prms = blocks_->at(chnenv_.CBuuid_).get_parameters();
-        group_index_ = std::stoi(blocks_->at(chnenv_.CBuuid_).get_parameter(prms[2]));
-        group_name_ = blocks_->at(chnenv_.CBuuid_).get_parameter(prms[4]);
-        group_comment_ = blocks_->at(chnenv_.CBuuid_).get_parameter(prms[6]);
+        CB_.parse(buffer_, blocks_->at(chnenv_.CBuuid_).get_parameters());
       }
 
       // extract associated CT data
       if ( blocks_->count(chnenv_.CTuuid_) == 1 )
       {
-        prms = blocks_->at(chnenv_.CTuuid_).get_parameters();
-        text_ = blocks_->at(chnenv_.CTuuid_).get_parameter(prms[4]) + std::string(" - ")
-              + blocks_->at(chnenv_.CTuuid_).get_parameter(prms[6]) + std::string(" - ")
-              + blocks_->at(chnenv_.CTuuid_).get_parameter(prms[8]);
-      }
-
-      // extract associated CD data
-      if ( blocks_->count(chnenv_.CDuuid_) == 1 )
-      {
-        prms = blocks_->at(chnenv_.CDuuid_).get_parameters();
-        xstepwidth_ = std::stod(blocks_->at(chnenv_.CDuuid_).get_parameter(prms[2]));
-        xunit_ = blocks_->at(chnenv_.CDuuid_).get_parameter(prms[5]);
-        // TODO
-        // xname_ = std::string("time");
-      }
-
-      // extract associated CP data
-      if ( blocks_->count(chnenv_.CPuuid_) == 1 )
-      {
-        prms = blocks_->at(chnenv_.CPuuid_).get_parameters();
-        num_bytes_ = std::stoi(blocks_->at(chnenv_.CPuuid_).get_parameter(prms[3]));
-        datatp_ = std::stoi(blocks_->at(chnenv_.CPuuid_).get_parameter(prms[4]));
-        signbits_ = std::stoi(blocks_->at(chnenv_.CPuuid_).get_parameter(prms[5]));
-        // byte_offset_ = std::stoul(blocks_->at(chnenv_.CPuuid_).get_parameter(prms[7]));
-      }
-
-      // extract associated Cb data
-      if ( blocks_->count(chnenv_.Cbuuid_) == 1 )
-      {
-        prms = blocks_->at(chnenv_.Cbuuid_).get_parameters();
-        buffer_offset_ = std::stoul(blocks_->at(chnenv_.Cbuuid_).get_parameter(prms[6]));
-        buffer_size_ = std::stoul(blocks_->at(chnenv_.Cbuuid_).get_parameter(prms[7]));
-        xoffset_ = std::stod(blocks_->at(chnenv_.Cbuuid_).get_parameter(prms[11]));
-      }
-
-      // extract associated CR data
-      if ( blocks_->count(chnenv_.CRuuid_) == 1 )
-      {
-        prms = blocks_->at(chnenv_.CRuuid_).get_parameters();
-        factor_ = std::stod(blocks_->at(chnenv_.CRuuid_).get_parameter(prms[3]));
-        offset_ = std::stod(blocks_->at(chnenv_.CRuuid_).get_parameter(prms[4]));
-        yunit_ = blocks_->at(chnenv_.CRuuid_).get_parameter(prms[7]);
+        CT_.parse(buffer_, blocks_->at(chnenv_.CTuuid_).get_parameters());
+        text_ = CT_.name_ + std::string(" - ")
+              + CT_.text_ + std::string(" - ")
+              + CT_.comment_;
       }
 
       // extract associated CN data
       if ( blocks_->count(chnenv_.CNuuid_) == 1 )
       {
-        prms = blocks_->at(chnenv_.CNuuid_).get_parameters();
-        name_ = blocks_->at(chnenv_.CNuuid_).get_parameter(prms[6]);
-        yname_ = name_;
-        comment_ = blocks_->at(chnenv_.CNuuid_).get_parameter(prms[8]);
-        // group_index_ = std::stoi(blocks_->at(chnenv_.CNuuid_).get_parameter(prms[2]));
+        CN_.parse(buffer_, blocks_->at(chnenv_.CNuuid_).get_parameters());
+	group_index_ = CN_.group_index_;
+	group_name_ = CN_.name_;
+	group_comment_ = CN_.comment_;
       }
 
-      // extract associated NO data
-      if ( blocks_->count(chnenv_.NOuuid_) == 1 )
+      if ( !chnenv_.compenv1_.uuid_.empty() && chnenv_.compenv2_.uuid_.empty() )
       {
-        prms = blocks_->at(chnenv_.NOuuid_).get_parameters();
-        origin_ = blocks_->at(chnenv_.NOuuid_).get_parameter(prms[4]);
-        origin_comment_ = blocks_->at(chnenv_.NOuuid_).get_parameter(prms[6]);
+        // normal dataset (single component)
+        // set common NT and CD keys if no others are specified
+        if (chnenv_.compenv1_.NTuuid_.empty()) chnenv_.compenv1_.NTuuid_ = chnenv_.NTuuid_;
+        if (chnenv_.compenv1_.CDuuid_.empty()) chnenv_.compenv1_.CDuuid_ = chnenv_.CDuuid_;
+
+        // comp_group1 contains y-data, x-data is based on xstepwidth_, xstart_ and the length of y-data
+        component_group comp_group1(chnenv_.compenv1_, blocks_, buffer_);
+        dimension_ = 1;
+
+        xstepwidth_ = comp_group1.CD_.dx_;
+        xunit_ = comp_group1.CD_.unit_;
+        ybuffer_offset_ = comp_group1.Cb_.offset_buffer_;
+        ybuffer_size_ = comp_group1.Cb_.number_bytes_;
+        xstart_ = comp_group1.Cb_.x0_;
+        yfactor_ = comp_group1.CR_.factor_;
+        yoffset_ = comp_group1.CR_.offset_;
+        yunit_ = comp_group1.CR_.unit_;
+        name_ = comp_group1.CN_.name_;
+        yname_ = comp_group1.CN_.name_;
+        comment_ = comp_group1.CN_.comment_;
+        ynum_bytes_ = comp_group1.CP_.bytes_;
+        ydatatp_ = comp_group1.CP_.numeric_type_;
+        ysignbits_ = comp_group1.CP_.signbits_;
+        // generate std::chrono::system_clock::time_point type
+        std::time_t ts = timegm(&comp_group1.NT_.tms_); // std::mktime(&tms);
+        trigger_time_ = std::chrono::system_clock::from_time_t(ts);
+        trigger_time_frac_secs_ = comp_group1.NT_.trigger_time_frac_secs_;
+        // calculate absolute trigger-time
+        addtime_ = static_cast<long int>(comp_group1.Cb_.add_time_);
+        absolute_trigger_time_ = trigger_time_ + std::chrono::seconds(addtime_);
+        //                                       + std::chrono::nanoseconds((long int)(trigger_time_frac_secs_*1.e9));
+      }
+      else if ( !chnenv_.compenv1_.uuid_.empty() && !chnenv_.compenv2_.uuid_.empty() )
+      {
+        // XY dataset (two components)
+        // set common NT and CD keys if no others are specified
+        if (chnenv_.compenv1_.NTuuid_.empty()) chnenv_.compenv1_.NTuuid_ = chnenv_.NTuuid_;
+        if (chnenv_.compenv1_.CDuuid_.empty()) chnenv_.compenv1_.CDuuid_ = chnenv_.CDuuid_;
+        if (chnenv_.compenv2_.NTuuid_.empty()) chnenv_.compenv2_.NTuuid_ = chnenv_.NTuuid_;
+        if (chnenv_.compenv2_.CDuuid_.empty()) chnenv_.compenv2_.CDuuid_ = chnenv_.CDuuid_;
+
+        // comp_group1 contains x-data, comp_group2 contains y-data
+        component_group comp_group1(chnenv_.compenv1_, blocks_, buffer_);
+        component_group comp_group2(chnenv_.compenv2_, blocks_, buffer_);
+        dimension_ = 2;
+
+        xbuffer_offset_ = comp_group2.Cb_.offset_buffer_;
+        xbuffer_size_ = comp_group2.Cb_.number_bytes_;
+        ybuffer_offset_ = comp_group1.Cb_.offset_buffer_;
+        ybuffer_size_ = comp_group1.Cb_.number_bytes_;
+        xfactor_ = comp_group2.CR_.factor_;
+        xoffset_ = comp_group2.CR_.offset_;
+        yfactor_ = comp_group1.CR_.factor_;
+        yoffset_ = comp_group1.CR_.offset_;
+        xdatatp_ = comp_group2.CP_.numeric_type_;
+        xsignbits_ = comp_group2.CP_.signbits_;
+        ydatatp_ = comp_group1.CP_.numeric_type_;
+        ysignbits_ = comp_group1.CP_.signbits_;
+        // generate std::chrono::system_clock::time_point type
+        std::time_t ts = timegm(&comp_group2.NT_.tms_); // std::mktime(&tms);
+        trigger_time_ = std::chrono::system_clock::from_time_t(ts);
+        trigger_time_frac_secs_ = comp_group2.NT_.trigger_time_frac_secs_;
+        absolute_trigger_time_ = trigger_time_;
+      }
+      else
+      {
+        // no datafield
       }
 
       // start converting binary buffer to imc::datatype
       if ( !chnenv_.CSuuid_.empty() ) convert_buffer();
+
+      // convert any non-UTF-8 codepage to UTF-8 and cleanse any text
+      convert_encoding();
+      cleanse_text();
     }
 
     // convert buffer to actual datatype
     void convert_buffer()
     {
-      // TODO no clue how/if/when to handle buffer offset/mask/subsequent_bytes
-      // etc. and whatever that shit is!
       std::vector<imc::parameter> prms = blocks_->at(chnenv_.CSuuid_).get_parameters();
       if ( prms.size() < 4)
       {
         throw std::runtime_error("CS block is invalid and features to few parameters");
       }
+
+      // extract (channel dependent) part of buffer
       unsigned long int buffstrt = prms[3].begin();
-      std::vector<unsigned char> CSbuffer( buffer_->begin()+buffstrt+1,
-                                           buffer_->begin()+buffstrt+buffer_size_+1 );
+      std::vector<unsigned char> yCSbuffer( buffer_->begin()+buffstrt+ybuffer_offset_+1,
+                                           buffer_->begin()+buffstrt+ybuffer_offset_+ybuffer_size_+1 );
 
       // determine number of values in buffer
-      unsigned long int num_values = CSbuffer.size()/(signbits_/8);
-      if ( num_values*(signbits_/8) != CSbuffer.size() )
+      unsigned long int ynum_values = (unsigned long int)(yCSbuffer.size()/(ysignbits_/8));
+      if ( ynum_values*(ysignbits_/8) != yCSbuffer.size() )
       {
-        throw std::runtime_error("CSbuffer and significant bits of datatype don't match");
+        throw std::runtime_error("CSbuffer and significant bits of y datatype don't match");
       }
 
-      // adjust size of ydata
-      ydata_.resize(num_values);
 
-      // distinguish numeric datatypes included in "imc_datatype"
-      if ( datatp_ == 1 )
+      if (dimension_ ==  1)
       {
-        imc::convert_data_to_type<imc_Ubyte>(CSbuffer,ydata_);
+        // process y-data
+        process_data(ydata_, ynum_values, ydatatp_, yCSbuffer);
+
+        // find appropriate precision for "xdata_" by means of "xstepwidth_"
+        xprec_ = (xstepwidth_ > 0 ) ? (int)ceil(fabs(log10(xstepwidth_))) : 10;
+
+        // fill xdata_
+        for ( unsigned long int i = 0; i < ynum_values; i++ )
+        {
+          xdata_.push_back(xstart_+(double)i*xstepwidth_);
+        }
       }
-      else if ( datatp_ == 2 )
+      else if (dimension_ == 2)
       {
-        imc::convert_data_to_type<imc_Sbyte>(CSbuffer,ydata_);
-      }
-      else if ( datatp_ == 3 )
-      {
-        imc::convert_data_to_type<imc_Ushort>(CSbuffer,ydata_);
-      }
-      else if ( datatp_ == 4 )
-      {
-        imc::convert_data_to_type<imc_Sshort>(CSbuffer,ydata_);
-      }
-      else if ( datatp_ == 5 )
-      {
-        imc::convert_data_to_type<imc_Ulongint>(CSbuffer,ydata_);
-      }
-      else if ( datatp_ == 6 )
-      {
-        imc::convert_data_to_type<imc_Slongint>(CSbuffer,ydata_);
-      }
-      else if ( datatp_ == 7 )
-      {
-        imc::convert_data_to_type<imc_float>(CSbuffer,ydata_);
-      }
-      else if ( datatp_ == 8 )
-      {
-        imc::convert_data_to_type<imc_double>(CSbuffer,ydata_);
-      }
-      // ...
-      else if ( datatp_ == 11 )
-      {
-        imc::convert_data_to_type<imc_digital>(CSbuffer,ydata_);
+        // process x- and y-data
+        std::vector<unsigned char> xCSbuffer( buffer_->begin()+buffstrt+xbuffer_offset_+1,
+                                            buffer_->begin()+buffstrt+xbuffer_offset_+xbuffer_size_+1 );
+
+        // determine number of values in buffer
+        unsigned long int xnum_values = (unsigned long int)(xCSbuffer.size()/(xsignbits_/8));
+        if ( xnum_values*(xsignbits_/8) != xCSbuffer.size() )
+        {
+          throw std::runtime_error("CSbuffer and significant bits of x datatype don't match");
+        }
+        if ( xnum_values != ynum_values )
+        {
+          throw std::runtime_error("x and y data have different number of values");
+        }
+
+        xprec_ = 9;
+
+        process_data(xdata_, xnum_values, xdatatp_, xCSbuffer);
+        process_data(ydata_, ynum_values, ydatatp_, yCSbuffer);
       }
       else
       {
-        throw std::runtime_error(std::string("unsupported/unknown datatype") + std::to_string(datatp_));
+        throw std::runtime_error("unsupported dimension");
       }
 
-      // fill xdata_
-      for ( unsigned long int i = 0; i < num_values; i++ )
-      {
-        xdata_.push_back(xoffset_+i*xstepwidth_);
-      }
+      transformData(xdata_, xfactor_, xoffset_);
+      transformData(ydata_, yfactor_, yoffset_);
+    }
 
-      // employ data transformation
-      if ( factor_ != 1.0 || offset_ != 0.0 )
+    // handle data type conversion
+    void process_data(std::vector<imc::datatype>& data_, size_t num_values, numtype datatp_, std::vector<unsigned char>& CSbuffer)
+    {
+      // adjust size of data
+      data_.resize(num_values);
+
+      // handle data type conversion
+      switch (datatp_)
       {
-        for ( imc::datatype& el: ydata_ )
-        {
-          // std::cout<<"value:"<<el.as_double()<<"\n";
-          el = imc::datatype(el.as_double()*factor_ + offset_);
+          case numtype::unsigned_byte:
+              imc::convert_data_to_type<imc_Ubyte>(CSbuffer, data_);
+              break;
+          case numtype::signed_byte:
+              imc::convert_data_to_type<imc_Sbyte>(CSbuffer, data_);
+              break;
+          case numtype::unsigned_short:
+              imc::convert_data_to_type<imc_Ushort>(CSbuffer, data_);
+              break;
+          case numtype::signed_short:
+              imc::convert_data_to_type<imc_Sshort>(CSbuffer, data_);
+              break;
+          case numtype::unsigned_long:
+              imc::convert_data_to_type<imc_Ulongint>(CSbuffer, data_);
+              break;
+          case numtype::signed_long:
+              imc::convert_data_to_type<imc_Slongint>(CSbuffer, data_);
+              break;
+          case numtype::ffloat:
+              imc::convert_data_to_type<imc_float>(CSbuffer, data_);
+              break;
+          case numtype::ddouble:
+              imc::convert_data_to_type<imc_double>(CSbuffer, data_);
+              break;
+          case numtype::two_byte_word_digital:
+              imc::convert_data_to_type<imc_digital>(CSbuffer, data_);
+              break;
+          case numtype::six_byte_unsigned_long:
+              imc::convert_data_to_type<imc_sixbyte>(CSbuffer, data_);
+              break;
+          default:
+              throw std::runtime_error(std::string("unsupported/unknown datatype ") + std::to_string(datatp_));
+      }
+    }
+
+    void transformData(std::vector<imc::datatype>& data, double factor, double offset) {
+        if (factor != 1.0 || offset != 0.0) {
+            for (imc::datatype& el : data) {
+                double fact = (factor == 0.0) ? 1.0 : factor;
+                el = imc::datatype(el.as_double() * fact + offset);
+            }
         }
+    }
+
+    // convert any description, units etc. to UTF-8 (by default)
+    void convert_encoding()
+    {
+      if ( !codepage_.empty() )
+      {
+        // construct iconv-compatible name for respective codepage
+        std::string cpn = std::string("CP") + codepage_;
+
+        // set up converter
+        std::string utf = std::string("UTF-8");
+        iconverter conv(cpn,utf);
+
+        conv.convert(name_);
+        conv.convert(comment_);
+        conv.convert(origin_);
+        conv.convert(origin_comment_);
+        conv.convert(text_);
+        conv.convert(language_code_);
+        conv.convert(yname_);
+        conv.convert(yunit_);
+        conv.convert(xname_);
+        conv.convert(xunit_);
+        conv.convert(group_name_);
+        conv.convert(group_comment_);
+      }
+    }
+
+    void cleanse_text()
+    {
+      escape_backslash(name_);
+      escape_backslash(comment_);
+      escape_backslash(origin_);
+      escape_backslash(origin_comment_);
+      escape_backslash(text_);
+      escape_backslash(language_code_);
+      escape_backslash(yname_);
+      escape_backslash(yunit_);
+      escape_backslash(xname_);
+      escape_backslash(xunit_);
+      escape_backslash(group_name_);
+      escape_backslash(group_comment_);
+    }
+
+    void escape_backslash(std::string &text)
+    {
+      char backslash = 0x5c;
+      std::string doublebackslash("\\\\");
+      for ( std::string::iterator it = text.begin(); it != text.end(); ++it )
+      {
+	if ( int(*it) == backslash ) {
+	  text.replace(it,it+1,doublebackslash);
+	  ++it;
+	}
       }
     }
 
     // get info string
     std::string get_info(int width = 20)
     {
+      // prepare printable trigger-time
+      std::time_t tt = std::chrono::system_clock::to_time_t(trigger_time_);
+      std::time_t att = std::chrono::system_clock::to_time_t(absolute_trigger_time_);
+
       std::stringstream ss;
       ss<<std::setw(width)<<std::left<<"uuid:"<<uuid_<<"\n"
         <<std::setw(width)<<std::left<<"name:"<<name_<<"\n"
         <<std::setw(width)<<std::left<<"comment:"<<comment_<<"\n"
         <<std::setw(width)<<std::left<<"origin:"<<origin_<<"\n"
+        <<std::setw(width)<<std::left<<"origin-comment:"<<origin_comment_<<"\n"
         <<std::setw(width)<<std::left<<"description:"<<text_<<"\n"
+        <<std::setw(width)<<std::left<<"trigger-time-nt:"<<std::put_time(std::gmtime(&tt),"%FT%T")<<"\n"
+        <<std::setw(width)<<std::left<<"trigger-time:"<<std::put_time(std::gmtime(&att),"%FT%T")<<"\n"
+        <<std::setw(width)<<std::left<<"language-code:"<<language_code_<<"\n"
+        <<std::setw(width)<<std::left<<"codepage:"<<codepage_<<"\n"
         <<std::setw(width)<<std::left<<"yname:"<<yname_<<"\n"
         <<std::setw(width)<<std::left<<"yunit:"<<yunit_<<"\n"
-        <<std::setw(width)<<std::left<<"datatype:"<<datatp_<<"\n"
-        <<std::setw(width)<<std::left<<"significant bits:"<<signbits_<<"\n"
-        <<std::setw(width)<<std::left<<"buffer-offset:"<<buffer_offset_<<"\n"
-        <<std::setw(width)<<std::left<<"buffer-size:"<<buffer_size_<<"\n"
+        <<std::setw(width)<<std::left<<"datatype:"<<ydatatp_<<"\n"
+        <<std::setw(width)<<std::left<<"significant bits:"<<ysignbits_<<"\n"
+        <<std::setw(width)<<std::left<<"buffer-offset:"<<ybuffer_offset_<<"\n"
+        <<std::setw(width)<<std::left<<"buffer-size:"<<ybuffer_size_<<"\n"
         <<std::setw(width)<<std::left<<"xname:"<<xname_<<"\n"
         <<std::setw(width)<<std::left<<"xunit:"<<xunit_<<"\n"
         <<std::setw(width)<<std::left<<"xstepwidth:"<<xstepwidth_<<"\n"
-        <<std::setw(width)<<std::left<<"xoffset:"<<xoffset_<<"\n"
-        <<std::setw(width)<<std::left<<"factor:"<<factor_<<"\n"
-        <<std::setw(width)<<std::left<<"offset:"<<offset_<<"\n"
+        <<std::setw(width)<<std::left<<"xoffset:"<<xstart_<<"\n"
+        <<std::setw(width)<<std::left<<"factor:"<<yfactor_<<"\n"
+        <<std::setw(width)<<std::left<<"offset:"<<yoffset_<<"\n"
         <<std::setw(width)<<std::left<<"group:"<<"("<<group_index_<<","<<group_name_
                                                     <<","<<group_comment_<<")"<<"\n"
-        <<std::setw(width)<<std::left<<"ydata:"<<imc::joinvec<imc::datatype>(ydata_)<<"\n"
-        <<std::setw(width)<<std::left<<"xdata:"<<imc::joinvec<double>(xdata_)<<"\n";
+        <<std::setw(width)<<std::left<<"ydata:"<<imc::joinvec<imc::datatype>(ydata_,6,9,true)<<"\n"
+        <<std::setw(width)<<std::left<<"xdata:"<<imc::joinvec<imc::datatype>(xdata_,6,xprec_,true)<<"\n";
         // <<std::setw(width)<<std::left<<"aff. blocks:"<<chnenv_.get_json()<<"\n";
       return ss.str();
     }
@@ -353,26 +699,36 @@ namespace imc
     // provide JSON string of metadata
     std::string get_json(bool include_data = false)
     {
+      // prepare printable trigger-time
+      std::time_t tt = std::chrono::system_clock::to_time_t(trigger_time_);
+      std::time_t att = std::chrono::system_clock::to_time_t(absolute_trigger_time_);
+
       std::stringstream ss;
       ss<<"{"<<"\"uuid\":\""<<uuid_
              <<"\",\"name\":\""<<name_
              <<"\",\"comment\":\""<<comment_
              <<"\",\"origin\":\""<<origin_
+             <<"\",\"origin-comment\":\""<<origin_comment_
              <<"\",\"description\":\""<<text_
-             <<"\",\"yname\":\""<<yname_
-             <<"\",\"yunit\":\""<<yunit_
-             <<"\",\"significantbits\":\""<<signbits_
-             <<"\",\"xname\":\""<<xname_
-             <<"\",\"xunit\":\""<<xunit_
+             <<"\",\"trigger-time-nt\":\""<<std::put_time(std::gmtime(&tt),"%FT%T")
+             <<"\",\"trigger-time\":\""<<std::put_time(std::gmtime(&att),"%FT%T")
+             <<"\",\"language-code\":\""<<language_code_
+             <<"\",\"codepage\":\""<<codepage_
+             <<"\",\"yname\":\""<<prepjsonstr(yname_)
+             <<"\",\"yunit\":\""<<prepjsonstr(yunit_)
+             <<"\",\"significantbits\":\""<<ysignbits_
+             <<"\",\"buffer-size\":\""<<ybuffer_size_
+             <<"\",\"xname\":\""<<prepjsonstr(xname_)
+             <<"\",\"xunit\":\""<<prepjsonstr(xunit_)
              <<"\",\"xstepwidth\":\""<<xstepwidth_
-             <<"\",\"xoffset\":\""<<xoffset_
+             <<"\",\"xoffset\":\""<<xstart_
              <<"\",\"group\":{"<<"\"index\":\""<<group_index_
                                <<"\",\"name\":\""<<group_name_
                                <<"\",\"comment\":\""<<group_comment_<<"\""<<"}";
       if ( include_data )
       {
-        ss<<",\"ydata\":"<<imc::joinvec<imc::datatype>(ydata_,0)
-          <<",\"xdata\":"<<imc::joinvec<double>(xdata_,0);
+        ss<<",\"ydata\":"<<imc::joinvec<imc::datatype>(ydata_,0,9,true)
+          <<",\"xdata\":"<<imc::joinvec<imc::datatype>(xdata_,0,xprec_,true);
       }
       // ss<<"\",\"aff. blocks\":\""<<chnenv_.get_json()
       ss<<"}";
@@ -380,8 +736,27 @@ namespace imc
       return ss.str();
     }
 
+    // prepare string value for usage in JSON dump
+    std::string prepjsonstr(std::string value)
+    {
+      std::stringstream ss;
+      ss<<quoted(value);
+      return strip_quotes(ss.str());
+    }
+
+    // remove any leading or trailing double quotes
+    std::string strip_quotes(std::string astring)
+    {
+      // head
+      if ( astring.front() == '"' ) astring.erase(astring.begin()+0);
+      // tail
+      if ( astring.back() == '"' ) astring.erase(astring.end()-1);
+
+      return astring;
+    }
+
     // print channel
-    void print(std::string filename, const char sep = ',', int width = 25)
+    void print(std::string filename, const char sep = ',', int width = 25, int yprec = 9)
     {
       std::ofstream fou(filename);
 
@@ -402,13 +777,16 @@ namespace imc
       {
         if ( sep == ' ' )
         {
-          fou<<std::setprecision(9)<<std::fixed
+          fou<<std::setprecision(xprec_)<<std::fixed
              <<std::setw(width)<<std::left<<xdata_[i]
+             <<std::setprecision(yprec)<<std::fixed
              <<std::setw(width)<<std::left<<ydata_[i]<<"\n";
         }
         else
         {
-          fou<<xdata_[i]<<sep<<ydata_[i]<<"\n";
+          fou<<std::setprecision(xprec_)<<std::fixed<<xdata_[i]
+             <<sep
+             <<std::setprecision(yprec)<<std::fixed<<ydata_[i]<<"\n";
         }
       }
 

lib/imc_conversion.hpp

@@ -1,6 +1,6 @@
 //---------------------------------------------------------------------------//
 
-#ifndef IMCCONVRSION
+#ifndef IMCCONVERSION
 #define IMCCONVERSION
 
 #include <vector>

lib/imc_datatype.hpp

@@ -16,12 +16,12 @@ namespace imc
   //
   // e.g. ARM Cortex-A72 armv7l gcc version 10.2.0 (Ubuntu 10.2.0-13ubuntu1)
   // #ifdef __arm__
-  typedef unsigned long int imc_Ulongint;
-  typedef signed long int imc_Slongint;
+  // typedef unsigned long int imc_Ulongint;
+  // typedef signed long int imc_Slongint;
   // e.g. Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz x86_64 gcc version 10.2.0 (Ubuntu 10.2.0-13ubuntu1)
   // #ifdef i386 __i386 __i386__
-  // typedef unsigned int imc_Ulongint;
-  // typedef signed int imc_Slongint;
+  typedef unsigned int imc_Ulongint;
+  typedef signed int imc_Slongint;
   //
   typedef float imc_float;
   typedef double imc_double;
@@ -30,7 +30,11 @@ namespace imc
   // typedef <whatever that is ->... > "imc Devices Transitional Recording"
   // typedf <sometimestamptype>  "Timestamp Ascii"
   typedef char16_t imc_digital;
-  // typedef <  > imc_sixbyte "6byte unsigned long"
+  //
+  typedef struct {
+      unsigned char bytes[6];
+  } imc_sixbyte;
+
 
   class datatype
   {
@@ -44,13 +48,14 @@ namespace imc
     imc_float sfloat_;     // 6
     imc_double sdouble_;   // 7
     imc_digital sdigital_; // 10
-    short int dtidx_;      // \in \{0,...,7,10\}
+    imc_sixbyte sixbyte_;  // 13
+    short int dtidx_;      // \in \{0,...,7,10,13\}
   public:
     datatype(): ubyte_(0), sbyte_(0),
                 ushort_(0), sshort_(0),
-                ulint_(0.0), slint_(0.0),
+                ulint_(0), slint_(0),
                 sfloat_(0.0), sdouble_(0.0),
-                sdigital_(0),
+                sdigital_(0), sixbyte_({0}),
                 dtidx_(0) { };
     // every supported datatype gets its own constructor
     datatype(imc_Ubyte num): ubyte_(num), dtidx_(0) {};
@@ -61,11 +66,14 @@ namespace imc
     datatype(imc_Slongint num): slint_(num), dtidx_(5) {};
     datatype(imc_float num): sfloat_(num), dtidx_(6) {};
     datatype(imc_double num): ubyte_(0), sbyte_(0), ushort_(0), sshort_(0),
-                              ulint_(0.0), slint_(0.0), sfloat_(0.0), sdouble_(num),
-                              sdigital_(0), dtidx_(7) {};
+                              ulint_(0), slint_(0), sfloat_(0.0), sdouble_(num),
+                              sdigital_(0), sixbyte_({0}), dtidx_(7) {};
     datatype(imc_digital num): ubyte_(0), sbyte_(0), ushort_(0), sshort_(0),
-                               ulint_(0.0), slint_(0.0), sfloat_(0.0), sdouble_(num),
-                               sdigital_(num), dtidx_(10) {};
+                               ulint_(0), slint_(0), sfloat_(0.0), sdouble_(0.0),
+                               sdigital_(num), sixbyte_({0}), dtidx_(10) {};
+    datatype(imc_sixbyte num): ubyte_(0), sbyte_(0), ushort_(0), sshort_(0),
+                               ulint_(0), slint_(0), sfloat_(0.0), sdouble_(0.0),
+                               sdigital_(0), sixbyte_(num), dtidx_(13) {};
 
     // identify type
     short int& dtype() { return dtidx_; }
@@ -82,6 +90,7 @@ namespace imc
       this->sfloat_ = num.sfloat_;
       this->sdouble_ = num.sdouble_;
       this->sdigital_ = num.sdigital_;
+      this->sixbyte_ = num.sixbyte_;
       this->dtidx_ = num.dtidx_;
     }
 
@@ -99,6 +108,7 @@ namespace imc
         this->sfloat_ = num.sfloat_;
         this->sdouble_ = num.sdouble_;
         this->sdigital_ = num.sdigital_;
+        this->sixbyte_ = num.sixbyte_;
         this->dtidx_ = num.dtidx_;
       }
 
@@ -160,6 +170,12 @@ namespace imc
       this->dtidx_ = 10;
       return *this;
     }
+    datatype& operator=(const imc_sixbyte &num)
+    {
+      this->sixbyte_ = num;
+      this->dtidx_ = 13;
+      return *this;
+    }
 
     // obtain number as double
     double as_double()
@@ -174,6 +190,13 @@ namespace imc
       else if ( dtidx_ == 6 ) num = (double)sfloat_;
       else if ( dtidx_ == 7 ) num = (double)sdouble_;
       else if ( dtidx_ == 10 ) num = static_cast<double>(sdigital_);
+      else if ( dtidx_ == 13 ) {
+        unsigned long long value = 0;
+        for (int i = 0; i < 6; ++i) {
+            value |= static_cast<unsigned long long>(sixbyte_.bytes[i]) << (8 * i);
+        }
+        num = static_cast<double>(value);
+      }
       return num;
     }
 
@@ -189,6 +212,13 @@ namespace imc
       else if ( num.dtidx_ == 6 ) out<<num.sfloat_;
       else if ( num.dtidx_ == 7 ) out<<num.sdouble_;
       else if ( num.dtidx_ == 10 ) out<<static_cast<double>(num.sdigital_);
+      else if ( num.dtidx_ == 13 ) {
+        unsigned long long value = 0;
+        for (int i = 0; i < 6; ++i) {
+            value |= static_cast<unsigned long long>(num.sixbyte_.bytes[i]) << (8 * i);
+        }
+        out<<static_cast<double>(value);
+      }
       return out;
     }
 

lib/imc_key.hpp

@@ -84,10 +84,12 @@ namespace imc
     // noncritical keys
     key(false,"NO","origin of data",1),
     key(false,"NT","timestamp of trigger",1),
+    key(false,"NT","timestamp of trigger",2),
     key(false,"ND","(color) display properties",1),
     key(false,"NU","user defined key",1),
     key(false,"Np","property of channel",1),
-    key(false,"NE","extraction rule for BUS channels",1)
+    key(false,"NE","extraction rule for BUS channels",1),
+    key(false,"NL","language info and code page",1)
 
   };
 

lib/imc_object.hpp

@@ -4,6 +4,7 @@
 #define IMCOBJECT
 
 #include <time.h>
+#include <math.h>
 #include "imc_key.hpp"
 
 //---------------------------------------------------------------------------//
@@ -245,7 +246,7 @@ namespace imc
     // construct members by parsing particular parameters from buffer
     void parse(const std::vector<unsigned char>* buffer, const std::vector<parameter>& parameters)
     {
-      if ( parameters.size() < 4 ) throw std::runtime_error("invalid number of parameters in CD2");
+      if ( parameters.size() < 4 ) throw std::runtime_error("invalid number of parameters in CC");
       component_index_ = std::stoi(get_parameter(buffer,&parameters[2]));
       analog_digital_ = ( std::stoi(get_parameter(buffer,&parameters[3])) == 2 );
     }
@@ -272,7 +273,9 @@ namespace imc
     imc_devices_transitional_recording,
     timestamp_ascii,
     two_byte_word_digital,
-    six_byte_unsigned_long
+    eight_byte_unsigned_long,
+    six_byte_unsigned_long,
+    eight_byte_signed_long
   };
 
   // packaging information of component (corresponds to key CP)
@@ -292,8 +295,8 @@ namespace imc
     {
       if ( parameters.size() < 10 ) throw std::runtime_error("invalid number of parameters in CP");
       buffer_reference_ = std::stoi(get_parameter(buffer,&parameters[2]));
-      numeric_type_ = (numtype)std::stoi(get_parameter(buffer,&parameters[3]));
-      bytes_ = std::stoi(get_parameter(buffer,&parameters[4]));
+      bytes_ = std::stoi(get_parameter(buffer,&parameters[3]));
+      numeric_type_ = (numtype)std::stoi(get_parameter(buffer,&parameters[4]));
       signbits_ = std::stoi(get_parameter(buffer,&parameters[5]));
       mask_ = std::stoi(get_parameter(buffer,&parameters[6]));
       offset_ = std::stoul(get_parameter(buffer,&parameters[7]));
@@ -336,7 +339,7 @@ namespace imc
     // construct members by parsing particular parameters from buffer
     void parse(const std::vector<unsigned char>* buffer, const std::vector<parameter>& parameters)
     {
-      if ( parameters.size() < 13 ) throw std::runtime_error("invalid number of parameters in CD2");
+      if ( parameters.size() < 13 ) throw std::runtime_error("invalid number of parameters in Cb");
       number_buffers_ = std::stoul(get_parameter(buffer,&parameters[2]));
       bytes_userinfo_ = std::stoul(get_parameter(buffer,&parameters[3]));
       buffer_reference_ = std::stoul(get_parameter(buffer,&parameters[4]));
@@ -378,7 +381,7 @@ namespace imc
     // construct members by parsing particular parameters from buffer
     void parse(const std::vector<unsigned char>* buffer, const std::vector<parameter>& parameters)
     {
-      if ( parameters.size() < 8 ) throw std::runtime_error("invalid number of parameters in CD2");
+      if ( parameters.size() < 8 ) throw std::runtime_error("invalid number of parameters in CR");
       transform_ = (get_parameter(buffer,&parameters[2]) == std::string("1"));
       factor_ = std::stod(get_parameter(buffer,&parameters[3]));
       offset_ = std::stod(get_parameter(buffer,&parameters[4]));
@@ -410,7 +413,7 @@ namespace imc
     // construct members by parsing particular parameters from buffer
     void parse(const std::vector<unsigned char>* buffer, const std::vector<parameter>& parameters)
     {
-      if ( parameters.size() < 9 ) throw std::runtime_error("invalid number of parameters in CD2");
+      if ( parameters.size() < 9 ) throw std::runtime_error("invalid number of parameters in CN");
       group_index_ = std::stoul(get_parameter(buffer,&parameters[2]));
       index_bit_ = (get_parameter(buffer,&parameters[4]) == std::string("1"));
       name_ = get_parameter(buffer,&parameters[6]);
@@ -439,7 +442,7 @@ namespace imc
     // construct members by parsing particular parameters from buffer
     void parse(const std::vector<unsigned char>* buffer, const std::vector<parameter>& parameters)
     {
-      if ( parameters.size() < 4 ) throw std::runtime_error("invalid number of parameters in CD2");
+      if ( parameters.size() < 4 ) throw std::runtime_error("invalid number of parameters in CS");
       index_ = std::stoul(get_parameter(buffer,&parameters[2]));
     }
 
@@ -454,6 +457,21 @@ namespace imc
     }
   };
 
+  // language (corresponds to key NL)
+  struct language
+  {
+    std::string codepage_;
+    std::string language_code_;
+
+    // construct members by parsing particular parameters from buffer
+    void parse(const std::vector<unsigned char>* buffer, const std::vector<parameter>& parameters)
+    {
+      if (parameters.size() < 4) throw std::runtime_error("invalid number of parameters in NL");
+      codepage_ = get_parameter(buffer, &parameters[2]);
+      language_code_ = get_parameter(buffer, &parameters[3]);
+    }
+  };
+
   // origin of data (corresponds to key NO)
   struct origin_data
   {
@@ -464,7 +482,7 @@ namespace imc
     // construct members by parsing particular parameters from buffer
     void parse(const std::vector<unsigned char>* buffer, const std::vector<parameter>& parameters)
     {
-      if ( parameters.size() < 7 ) throw std::runtime_error("invalid number of parameters in CD2");
+      if ( parameters.size() < 7 ) throw std::runtime_error("invalid number of parameters in NO");
       origin_ = ( get_parameter(buffer,&parameters[2]) == std::string("1") );
       generator_ = get_parameter(buffer,&parameters[4]);
       comment_ = get_parameter(buffer,&parameters[6]);
@@ -484,45 +502,30 @@ namespace imc
   // trigger timestamp (corresponds to key NT1)
   struct triggertime
   {
-    int day_, month_, year_;
-    int hour_, minute_;
-    double second_;
-    std::string timestamp_;
+    std::tm tms_;
+    double trigger_time_frac_secs_;
 
     // construct members by parsing particular parameters from buffer
     void parse(const std::vector<unsigned char>* buffer, const std::vector<parameter>& parameters)
     {
-      if ( parameters.size() < 8 ) throw std::runtime_error("invalid number of parameters in CD2");
-      day_ = std::stoi( get_parameter(buffer,&parameters[2]) );
-      month_ = std::stoi( get_parameter(buffer,&parameters[3]) );
-      year_ = std::stoi( get_parameter(buffer,&parameters[4]) );
-      hour_ = std::stoi( get_parameter(buffer,&parameters[5]) );
-      minute_ = std::stoi( get_parameter(buffer,&parameters[6]) );
-      second_ = std::stod( get_parameter(buffer,&parameters[7]) );
-
-      time_t rawtime;
-      struct tm ts;
-      time(&rawtime);
-      localtime_r(&rawtime,&ts);
-      ts.tm_mday = day_;
-      ts.tm_mon = month_-1;
-      ts.tm_year = year_-1900;
-      ts.tm_hour = hour_;
-      ts.tm_min = minute_;
-      ts.tm_sec = (int)second_;
-      asctime_r(&ts,&timestamp_[0]);
-      // timestamp_ = std::to_string(year_) + std::string("-") + std::to_string(month_)
-      //                                    + std::string("-") + std::to_string(day_)
-      //                                    + std::string("T") + std::to_string(hour_)
-      //                                    + std::string(":") + std::to_string(minute_)
-      //                                    + std::string(":") + std::to_string(second_);
+      if ( parameters.size() < 8 ) throw std::runtime_error("invalid number of parameters in NT1");
+      tms_ = std::tm();
+      tms_.tm_mday = std::stoi( get_parameter(buffer,&parameters[2]) );
+      tms_.tm_mon = std::stoi( get_parameter(buffer,&parameters[3]) ) - 1;
+      tms_.tm_year = std::stoi( get_parameter(buffer,&parameters[4]) ) - 1900;
+      tms_.tm_hour = std::stoi( get_parameter(buffer,&parameters[5]) );
+      tms_.tm_min = std::stoi( get_parameter(buffer,&parameters[6]) );
+      long double secs = std::stold( get_parameter(buffer,&parameters[7]) );
+      double secs_int;
+      trigger_time_frac_secs_ = modf((double)secs,&secs_int);
+      tms_.tm_sec = (int)secs_int;
     }
 
     // get info string
     std::string get_info(int width = 20)
     {
       std::stringstream ss;
-      ss<<std::setw(width)<<std::left<<"timestamp:"<<timestamp_<<"\n";
+      ss<<std::setw(width)<<std::left<<"timestamp:"<<std::put_time(&tms_, "%Y-%m-%dT%H:%M:%S")<<"\n";
       return ss.str();
     }
   };

lib/imc_raw.hpp

@@ -58,6 +58,8 @@ namespace imc
     // open file and stream data into buffer
     void fill_buffer()
     {
+      buffer_.clear();
+
       // open file and put data in buffer
       try {
         std::ifstream fin(raw_file_.c_str(),std::ifstream::binary);
@@ -76,6 +78,8 @@ namespace imc
     // parse all raw blocks in buffer
     void parse_blocks()
     {
+      rawblocks_.clear();
+
       // reset counter to identify computational complexity
       cplxcnt_ = 0;
 
@@ -123,20 +127,20 @@ namespace imc
                   leng.push_back((char)*(it+pos));
                   pos++;
                 }
-                unsigned long length = std::stoul(leng);
+                unsigned long int length = std::stoul(leng);
 
                 // declare and initialize corresponding key and block
                 // imc::key bkey( *(it+1)==imc::key_crit_ , newkey,
                 //                imc::keys.at(newkey).description_, version );
-                imc::block blk(itkey,it-buffer_.begin(),
-                                     it-buffer_.begin()+pos+1+length,
+                imc::block blk(itkey,(unsigned long int)(it-buffer_.begin()),
+                                     (unsigned long int)(it-buffer_.begin()+pos+1+length),
                                      raw_file_, &buffer_);
 
                 // add block to list
                 rawblocks_.push_back(blk);
 
                 // skip the remaining block according to its length
-                if ( it-buffer_.begin()+length < buffer_.size() )
+                if ( (unsigned long int)(it-buffer_.begin()+length) < (unsigned long int)(buffer_.size()) )
                 {
                   std::advance(it,length);
                 }
@@ -190,6 +194,8 @@ namespace imc
     // generate map of blocks using their uuid
     void generate_block_map()
     {
+      mapblocks_.clear();
+
       for ( imc::block blk: rawblocks_ )
       {
         mapblocks_.insert( std::pair<std::string,imc::block>(blk.get_uuid(),blk) );
@@ -199,35 +205,82 @@ namespace imc
     // generate channel "environments"
     void generate_channel_env()
     {
+      channels_.clear();
+
       // declare single channel environment
       imc::channel_env chnenv;
       chnenv.reset();
 
+      imc::component_env *compenv_ptr = nullptr;
+
       // collect affiliate blocks for every channel WITH CHANNEL and AFFILIATE
       // BLOCK CORRESPONDENCE GOVERNED BY BLOCK ORDER IN BUFFER!!
       for ( imc::block blk: rawblocks_ )
       {
+        if ( blk.get_key().name_ == "NO" ) chnenv.NOuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "NL" ) chnenv.NLuuid_ = blk.get_uuid();
 
-        if ( blk.get_key().name_ == "CN" ) chnenv.CNuuid_ = blk.get_uuid();
-        else if ( blk.get_key().name_ == "CD" ) chnenv.CDuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "CB" ) chnenv.CBuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "CG" ) chnenv.CGuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "CI" ) chnenv.CIuuid_ = blk.get_uuid();
         else if ( blk.get_key().name_ == "CT" ) chnenv.CTuuid_ = blk.get_uuid();
-        else if ( blk.get_key().name_ == "Cb" ) chnenv.Cbuuid_ = blk.get_uuid();
-        else if ( blk.get_key().name_ == "CP" ) chnenv.CPuuid_ = blk.get_uuid();
-        else if ( blk.get_key().name_ == "CR" ) chnenv.CRuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "CN" ) chnenv.CNuuid_ = blk.get_uuid();
         else if ( blk.get_key().name_ == "CS" ) chnenv.CSuuid_ = blk.get_uuid();
-        else if ( blk.get_key().name_ == "NT" ) chnenv.NTuuid_ = blk.get_uuid();
-        else if ( blk.get_key().name_ == "NO" ) chnenv.NOuuid_ = blk.get_uuid();
+
+        else if ( blk.get_key().name_ == "CC" )
+        {
+          // a new component group is started
+          // TODO: can we avoid to parse the whole component here?
+          imc::component component;
+          component.parse(&buffer_, blk.get_parameters());
+          if ( component.component_index_ == 1 ) compenv_ptr = &chnenv.compenv1_;
+          else if ( component.component_index_ == 2 ) compenv_ptr = &chnenv.compenv2_;
+          else throw std::runtime_error("invalid component index in CC block");
+          compenv_ptr->CCuuid_ = blk.get_uuid();
+          compenv_ptr->uuid_ = compenv_ptr->CCuuid_;
+        }
+        else if ( blk.get_key().name_ == "CD" )
+        {
+          if (compenv_ptr == nullptr) chnenv.CDuuid_ = blk.get_uuid();
+          else compenv_ptr->CDuuid_ = blk.get_uuid();
+        }
+        else if ( blk.get_key().name_ == "NT" )
+        {
+          if (compenv_ptr == nullptr) chnenv.NTuuid_ = blk.get_uuid();
+          else compenv_ptr->NTuuid_ = blk.get_uuid();
+        }
+        else if ( blk.get_key().name_ == "Cb" ) compenv_ptr->Cbuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "CP" ) compenv_ptr->CPuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "CR" ) compenv_ptr->CRuuid_ = blk.get_uuid();
+
 
         // check for currently associated channel
+        // TODO: CNuuid is not unique for multichannel data
         if ( !chnenv.CNuuid_.empty() )
         {
-          // any component/channel is closed by any of {CS, CC, CG, CB}
-          if ( blk.get_key().name_ == "CS" || blk.get_key().name_ == "CC"
-            || blk.get_key().name_ == "CG" || blk.get_key().name_ == "CB" )
+          // at the moment only a single channel is supported
+          // any channel is closed by any of {CB, CG, CI, CT, CS}
+          if ( blk.get_key().name_ == "CB" || blk.get_key().name_ == "CG"
+            || blk.get_key().name_ == "CI" || blk.get_key().name_ == "CT"
+            || blk.get_key().name_ == "CS" )
           {
             // provide UUID for channel
+            // for multi component channels exactly one CN is available
             chnenv.uuid_ = chnenv.CNuuid_;
 
+            // for multichannel data there may be multiple channels referring to
+            // the same (final) CS block (in contrast to what the IMC software
+            // documentation seems to suggest) resulting in all channels missing
+            // a CS block except for the very last
+            if ( chnenv.CSuuid_.empty() ) {
+              for ( imc::block blkCS: rawblocks_ ) {
+                if ( blkCS.get_key().name_ == "CS"
+                  && blkCS.get_begin() > (unsigned long int)stol(chnenv.uuid_) ) {
+                  chnenv.CSuuid_ = blkCS.get_uuid();
+                }
+              }
+            }
+
             // create channel object and add it to the map of channels
             channels_.insert( std::pair<std::string,imc::channel>
               (chnenv.CNuuid_,imc::channel(chnenv,&mapblocks_,&buffer_))
@@ -235,6 +288,14 @@ namespace imc
 
             // reset channel uuid
             chnenv.CNuuid_.clear();
+
+            chnenv.CBuuid_.clear();
+            chnenv.CGuuid_.clear();
+            chnenv.CIuuid_.clear();
+            chnenv.CTuuid_.clear();
+            chnenv.CSuuid_.clear();
+
+            compenv_ptr = nullptr;
           }
         }
 
@@ -242,17 +303,17 @@ namespace imc
         // already belong to NEXT component
         if ( blk.get_key().name_ == "CB" ) chnenv.CBuuid_ = blk.get_uuid();
         else if ( blk.get_key().name_ == "CG" ) chnenv.CGuuid_ = blk.get_uuid();
-        else if ( blk.get_key().name_ == "CC" ) chnenv.CCuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "CI" ) chnenv.CIuuid_ = blk.get_uuid();
+        else if ( blk.get_key().name_ == "CT" ) chnenv.CTuuid_ = blk.get_uuid();
       }
     }
 
-
   public:
 
     // provide buffer size
     unsigned long int buffer_size()
     {
-      return buffer_.size();
+      return (unsigned long int)buffer_.size();
     }
 
     // get blocks
@@ -376,7 +437,7 @@ namespace imc
         std::filesystem::path pf = pd / filenam;
 
         // and print the channel
-        it->second.print(pf,sep);
+        it->second.print(pf.u8string(),sep);
       }
     }
 

makefile

@@ -8,22 +8,25 @@ EXE = imctermite
 # directory names
 SRC = src/
 LIB = lib/
-CYT = cython/
 PYT = python/
 
-# list headers
+# list headers and include directories
 HPP = $(wildcard $(LIB)/*.hpp)
+IPP = $(shell find $(LIB) -type f -name '*.hpp')
+KIB = $(shell find $(LIB) -type d)
+MIB = $(foreach dir,$(KIB),-I $(dir))
 
 # choose compiler and its options
 CC = g++ -std=c++17
-#OPT = -O3 -Wall -mavx -mno-tbm -mf16c -mno-f16c
-OPT = -O3 -Wall -Werror -Wunused-variable -Wsign-compare
+OPT = -O3 -Wall -Wconversion -Wpedantic -Werror -Wunused-variable -Wsign-compare
 
 # determine git version/commit and release tag
-GTAG := $(shell git tag | tail -n1)
+GTAG := $(shell git tag -l --sort=version:refname | tail -n1 | sed "s/$^v//g")
 GHSH := $(shell git rev-parse HEAD | head -c8)
-RTAG := v$(shell cat pip/setup.py | grep version | grep -oP "([0-9]\.){2}[0-9]")
-CTAG := v$(shell cat cython/setup.py | grep version | grep -oP "([0-9]\.){2}[0-9]")
+GVSN := $(shell cat python/VERSION | tr -d ' \n')
+
+# current timestamp
+TMS = $(shell date +%Y%m%dT%H%M%S)
 
 # define install location
 INST := /usr/local/bin
@@ -31,50 +34,50 @@ INST := /usr/local/bin
 #-----------------------------------------------------------------------------#
 # C++ and CLI tool
 
-# build exectuable
-$(EXE) : check-vtag $(RTAG) main.o
+# build executable
+$(EXE): check-tags $(GVSN) main.o
 	$(CC) $(OPT) main.o -o $@
 
 # build main.cpp and include git version/commit tag
-main.o : src/main.cpp $(HPP)
+main.o: src/main.cpp $(IPP)
 	@cp $< $<.cpp
 	@sed -i 's/TAGSTRING/$(GTAG)/g' $<.cpp
 	@sed -i 's/HASHSTRING/$(GHSH)/g' $<.cpp
-	$(CC) -c $(OPT) -I $(LIB) $<.cpp -o $@
+	@sed -i 's/TIMESTAMPSTRING/$(TMS)/g' $<.cpp
+	$(CC) -c $(OPT) $(MIB) $<.cpp -o $@
 	@rm $<.cpp
 
-install : $(EXE)
+install: $(EXE)
 	cp $< $(INST)/
 
-uninstall : $(INST)/$(EXE)
+uninstall: $(INST)/$(EXE)
 	rm $<
 
-cpp-clean :
+cpp-clean:
 	rm -vf $(EXE)
 	rm -vf *.o
 
 #-----------------------------------------------------------------------------#
-# linter and code check
+# C++ linter
 
 check-code:
 	cppcheck --enable=all -I lib/ src/main.cpp
 
 #-----------------------------------------------------------------------------#
-# check version consistency of git tags and version string in package.json
+# versions
 
-$(GTAG) :
+$(GTAG):
 	@echo "consistent versions check successful: building $(GTAG)"
 
-check-vtag:
-	@echo "git tag version:  "$(GTAG)
-	@echo "git commit hash:  "$(GHSH)
-	@echo "release version:  "$(RTAG)
-	@echo "module version:   "$(CTAG)
+check-tags:
+	@echo "latest git tag:  $(GTAG)"
+	@echo "latest git hash: $(GHSH)"
+	@echo "python version:  $(GVSN)"
 
 #-----------------------------------------------------------------------------#
 # Docker
 
-docker-build :
+docker-build:
 	docker build ./ --tag imctermite:0.1
 
 docker-run:
@@ -83,28 +86,29 @@ docker-run:
 #-----------------------------------------------------------------------------#
 # python
 
-cython-build : check-vtag $(CTAG) $(CYT)setup.py $(CYT)imc_termite.pxd $(CYT)py_imc_termite.pyx $(HPP)
-	python3 $(CYT)setup.py build_ext --inplace
-	cp -v imc_termite.cpython-*.so $(PYT)
+python-build: check-tags $(GVSN)
+	make -C python/ build-inplace
+	cp python/imctermite*.so ./ -v
 
-cython-install : check-vtag $(CTAG) $(CYT)setup.py $(CYT)imc_termite.pxd $(CYT)py_imc_termite.pyx $(HPP)
-	python3 $(CYT)setup.py install --record files_imctermite.txt
+python-clean:
+	make -C python/ clean
+	rm -vf imctermite*.so
 
-cython-clean :
-	rm -vf imc_termite.cpython-*.so
-	rm -vf $(PYT)imc_termite.cpython-*.so
-	rm -rvf build/
-
-#-----------------------------------------------------------------------------#
-# pip
-
-pip-release: check-vtag $(RTAG) cython-build
-	cd ./pip/ && make publish
+python-test:
+	PYTHONPATH=./ python python/examples/usage.py
 
 #-----------------------------------------------------------------------------#
 # clean
 
-clean: cpp-clean cython-clean
-	cd ./pip/ && make clean
+clean: cpp-clean python-clean
+
+#-----------------------------------------------------------------------------#
+# github actions
+
+github-action-lint: .github/workflows/pypi-deploy.yml
+	actionlint $<
+
+# for reference, see:
+# https://github.com/rhysd/actionlint
 
 #-----------------------------------------------------------------------------#

parquet/parquet/generate_deps.py

@@ -1,133 +0,0 @@
-#-----------------------------------------------------------------------------#
-
-import argparse
-import os
-
-#-----------------------------------------------------------------------------#
-
-parser = argparse.ArgumentParser(description='List all source dependencies')
-#parser.add_argument('pathToRepo',type=str,help='path of source repository')
-parser.add_argument('mainSource',type=str,help='main source file')
-parser.add_argument('depFile',type=str,help='file listing all dependencies')
-args = parser.parse_args()
-
-libpaths = ["/home/mario/Desktop/arrow/cpp/src/",
-            "/home/mario/Desktop/arrow/cpp/thrift_ep-install/include/",
-            "/home/mario/Desktop/arrow/cpp/boost_ep-prefix/src/boost_ep/"]
-
-#-----------------------------------------------------------------------------#
-
-def find_dependencies(srcfile, recdepth, cdeplist) :
-    """
-    Given a source file and its dependencies in the given repository path
-    list all further dependencies recursively
-
-    Args:
-        srcfile (string): path/name of source file
-        recdepth (integer): current recursion depth
-        cdeplist (list): current list of dependencies
-    Return:
-        deps (list): list of source files in repository, the source file depends on
-    """
-    # define indentation to visual recursion
-    indent = recdepth*("  ")
-
-    print("\n" + indent + "find_dependencies:"
-        + "\n" + indent + "1: " + srcfile
-        + "\n" + indent + "2: " + str(recdepth)
-        + "\n" + indent + "3: " + str(len(cdeplist)) + "\n")
-
-    # show dependencies so far
-    #print(cdeplist)
-
-    # generate dependencies by means of g++
-    libdeps = (" -I ").join(libpaths)
-    cmd = "g++ -c -MMD " + srcfile + " -I " + libdeps
-    print(indent + cmd )
-    os.system(cmd)
-
-    # open dependency file and extract list of sources
-    basename = srcfile.split('/')[-1].split('.')[0]
-    depfile = basename + '.d'
-    print(indent + "reading dependency file " + depfile)
-    with open(depfile,'r') as fin :
-        depslist = fin.readlines()
-
-    # delete dependencies and object files
-    os.system("rm " + basename + ".d")
-    os.system("rm " + basename + ".o")
-
-    # remove first line
-    depslist = depslist[1:]
-
-    # delete leading space and trailing backslash
-    depslistcl = [dep.lstrip(' ').rstrip(' \\\n') for dep in depslist]
-
-    # collect dependencies
-    newdeps = []
-
-    # check all dependencies recursively and collect further dependencies
-    count = 0
-    for dep in depslistcl :
-        # append source itself to list
-        if dep not in cdeplist :
-            print(indent + "adding dependency " + dep)
-            newdeps.append(dep)
-            count = count + 1
-    print(indent + "=> added " + str(count) + "/" + str(len(depslistcl)) )
-
-    # check recursion depth
-    if recdepth < 20 :
-        # check all dependencies of every single dependency
-        for dep in depslistcl :
-            # try to find corresponding *.cc, (*.cpp) file
-            depcc = dep.split('.')[0] + '.cc'
-            print(indent + "checking for " + depcc)
-            if os.path.exists(depcc) :
-                if depcc not in cdeplist and depcc not in newdeps :
-                    # add file itself as dependency
-                    newdeps.append(depcc)
-                    # find dependencies of single source
-                    newrecdeps = find_dependencies(depcc,recdepth+1,cdeplist+newdeps)
-                    # append to list
-                    for el in newrecdeps :
-                        if el not in newdeps :
-                            newdeps.append(el)
-                else :
-                    print(indent + "already in list")
-            else :
-                print(indent + "does not exist")
-
-    print("\n")
-
-    # provide list of dependencies
-    return newdeps
-
-#-----------------------------------------------------------------------------#
-
-if __name__== "__main__":
-
-    print("\nCLI arguments:\n" + str(args) + "\n")
-
-    # collect list of dependencies
-    deps = []
-
-    # start recursion with given source file
-    deps = find_dependencies(args.mainSource,0,[])
-
-    print("\nfinal list of dependencies: (" + str(len(deps)) + ")\n")
-    print(deps)
-    print("\n")
-
-    # remove any duplicates
-    depsuni = set(deps)
-    print("\nfinal set of dependencies: (" + str(len(depsuni)) + ")\n")
-    print(depsuni)
-    print("\n")
-
-    # write list of dependencies
-    with open(args.depFile,'w') as fout :
-        for el in depsuni :
-            fout.write(str(el) + '\n')
-
-#-----------------------------------------------------------------------------#

parquet/parquet/generate_make.py

@@ -1,22 +0,0 @@
-#-----------------------------------------------------------------------------#
-
-from pathlib import Path
-
-# find source files
-srcpaths = Path("src/").rglob('*.cc')
-deps =[ str(path) for path in srcpaths ]
-
-print(deps)
-
-with open('makefileobj','w') as fout :
-    for el in deps :
-        basnam = el.split('/')[-1]
-        print(str(el) + " : " + str(basnam) + " : " + str(basnam.split('.')[1]))
-        if basnam.split('.')[1] == 'cc' :
-            objfile = 'bin/' + basnam.replace('.cc','.o')
-            fout.write(objfile + " : " + el + "\n")
-            fout.write("\t" + "$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@\n")
-            fout.write("\n")
-
-
-#-----------------------------------------------------------------------------#

parquet/parquet/makefile

@@ -1,356 +0,0 @@
-#-----------------------------------------------------------------------------#
-
-CPP := g++ -std=c++14
-CPPFLAGS := -Woverflow -Wpedantic -Wextra -Waddress -Waligned-new -Walloc-zero
-
-SRC := src/
-BIN := bin/
-
-LIBS := -I src/src/ -I src/thrift_ep-install/include/ -I src/boost_ep-prefix/src/boost_ep/
-
-#-----------------------------------------------------------------------------#
-# prepare source
-#
-# before: $ cd arrow/cpp/  and compile relevant sources by
-#         $ cmake . -D ARROW_PARQUET=ON -D PARQUET_BUILD_EXAMPLES=ON -D ARROW_WITH_SNAPPY=ON
-#         $ cmake .. -D ARROW_PARQUET=ON ARROW_BUILD_EXAMPLES=ON
-
-lib :
-	cmake . -D ARROW_WITH_BROTLI=ON -D ARROW_WITH_BZ2=ON -D ARROW_WITH_LZ4=ON -D ARROW_WITH_SNAPPY=ON -D ARROW_WITH_ZLIB=ON -D ARROW_PARQUET=ON -D ARROW_PYTHON=ON
-
-# cp-src : deps.log
-# 	./src_copy.sh
-
-deps.log :
-	python3 generate_deps.py reader-writer.cc $@
-
-SRC := $(shell find $(SRC) -name '*.cc')
-
-# OBJ := $(apprefix obj/, $(SRC:%.cc=%.o))
-OBJ := $(addprefix $(BIN),$(notdir $(SRC:%.cc=%.o)))
-
-check :
-	@echo $(SRC)
-	@echo $(OBJ)
-
-# vpath %.cc src/
-
-reader-writer-example : reader-writer.cc $(OBJ) bin/utilmemory.o
-	$(CPP) $(CPPFLAGS) $< $(LIBS) -o $@ $(OBJ) bin/utilmemory.o
-
-# $(OBJ) : $(SRC)
-# 	$(CPP) $(OPT) -c $< -o $@ -I src/src/
-#
-# $(BIN)%.o : $(SRC)
-# 	$(CPP) $(OPT) -c $< -I src/src/ -o $@
-
-clean-obj :
-	rm -f $(OBJ)
-
-# => do build with cmake like here
-# https://arrow.apache.org/docs/developers/python.html#build-and-test
-
-#-----------------------------------------------------------------------------#
-
-bin/type.o : src/src/arrow/type.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/result.o : src/src/arrow/result.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder.o : src/src/arrow/builder.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/tensor.o : src/src/arrow/tensor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/table.o : src/src/arrow/table.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/extension_type.o : src/src/arrow/extension_type.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/device.o : src/src/arrow/device.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/memory_pool.o : src/src/arrow/memory_pool.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/datum.o : src/src/arrow/datum.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/record_batch.o : src/src/arrow/record_batch.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/compare.o : src/src/arrow/compare.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/visitor.o : src/src/arrow/visitor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/chunked_array.o : src/src/arrow/chunked_array.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/status.o : src/src/arrow/status.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/pretty_print.o : src/src/arrow/pretty_print.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/sparse_tensor.o : src/src/arrow/sparse_tensor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/buffer.o : src/src/arrow/buffer.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/scalar.o : src/src/arrow/scalar.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/string.o : src/src/arrow/util/string.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/utilmemory.o : src/src/arrow/util/memory.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/future.o : src/src/arrow/util/future.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/iterator.o : src/src/arrow/util/iterator.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/compression.o : src/src/arrow/util/compression.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/utf8.o : src/src/arrow/util/utf8.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/time.o : src/src/arrow/util/time.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/cpu_info.o : src/src/arrow/util/cpu_info.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/thread_pool.o : src/src/arrow/util/thread_pool.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bit_util.o : src/src/arrow/util/bit_util.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/logging.o : src/src/arrow/util/logging.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/basic_decimal.o : src/src/arrow/util/basic_decimal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/decimal.o : src/src/arrow/util/decimal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bit_block_counter.o : src/src/arrow/util/bit_block_counter.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/key_value_metadata.o : src/src/arrow/util/key_value_metadata.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/int_util.o : src/src/arrow/util/int_util.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/io_util.o : src/src/arrow/util/io_util.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bitmap_ops.o : src/src/arrow/util/bitmap_ops.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bitmap_builders.o : src/src/arrow/util/bitmap_builders.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bit_run_reader.o : src/src/arrow/util/bit_run_reader.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/value_parsing.o : src/src/arrow/util/value_parsing.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/string_builder.o : src/src/arrow/util/string_builder.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/formatting.o : src/src/arrow/util/formatting.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_primitive.o : src/src/arrow/array/array_primitive.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_dict.o : src/src/arrow/array/array_dict.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_binary.o : src/src/arrow/array/builder_binary.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_union.o : src/src/arrow/array/builder_union.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/concatenate.o : src/src/arrow/array/concatenate.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_nested.o : src/src/arrow/array/array_nested.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_decimal.o : src/src/arrow/array/array_decimal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_primitive.o : src/src/arrow/array/builder_primitive.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/data.o : src/src/arrow/array/data.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/diff.o : src/src/arrow/array/diff.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_nested.o : src/src/arrow/array/builder_nested.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_decimal.o : src/src/arrow/array/builder_decimal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_dict.o : src/src/arrow/array/builder_dict.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_binary.o : src/src/arrow/array/array_binary.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_adaptive.o : src/src/arrow/array/builder_adaptive.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_base.o : src/src/arrow/array/array_base.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/validate.o : src/src/arrow/array/validate.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_base.o : src/src/arrow/array/builder_base.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/util.o : src/src/arrow/array/util.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/caching.o : src/src/arrow/io/caching.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/memory.o : src/src/arrow/io/memory.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/interfaces.o : src/src/arrow/io/interfaces.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/buffered.o : src/src/arrow/io/buffered.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/file.o : src/src/arrow/io/file.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/strtod.o : src/src/arrow/vendored/double-conversion/strtod.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bignum.o : src/src/arrow/vendored/double-conversion/bignum.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/fixed-dtoa.o : src/src/arrow/vendored/double-conversion/fixed-dtoa.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/fast-dtoa.o : src/src/arrow/vendored/double-conversion/fast-dtoa.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/diy-fp.o : src/src/arrow/vendored/double-conversion/diy-fp.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/double-conversion.o : src/src/arrow/vendored/double-conversion/double-conversion.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bignum-dtoa.o : src/src/arrow/vendored/double-conversion/bignum-dtoa.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/cached-powers.o : src/src/arrow/vendored/double-conversion/cached-powers.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/api_aggregate.o : src/src/arrow/compute/api_aggregate.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/exec.o : src/src/arrow/compute/exec.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/kernel.o : src/src/arrow/compute/kernel.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/registry.o : src/src/arrow/compute/registry.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/function.o : src/src/arrow/compute/function.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/cast.o : src/src/arrow/compute/cast.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/api_vector.o : src/src/arrow/compute/api_vector.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/api_scalar.o : src/src/arrow/compute/api_scalar.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/codegen_internal.o : src/src/arrow/compute/kernels/codegen_internal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/column_scanner.o : src/src/parquet/column_scanner.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/statistics.o : src/src/parquet/statistics.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/internal_file_decryptor.o : src/src/parquet/internal_file_decryptor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/column_writer.o : src/src/parquet/column_writer.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/encryption.o : src/src/parquet/encryption.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/file_reader.o : src/src/parquet/file_reader.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/properties.o : src/src/parquet/properties.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/encryption_internal.o : src/src/parquet/encryption_internal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/internal_file_encryptor.o : src/src/parquet/internal_file_encryptor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/types.o : src/src/parquet/types.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/encoding.o : src/src/parquet/encoding.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/metadata.o : src/src/parquet/metadata.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/printer.o : src/src/parquet/printer.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/level_conversion.o : src/src/parquet/level_conversion.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/deprecated_io.o : src/src/parquet/deprecated_io.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/file_writer.o : src/src/parquet/file_writer.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/schema.o : src/src/parquet/schema.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/platform.o : src/src/parquet/platform.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/column_reader.o : src/src/parquet/column_reader.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@

parquet/parquet/makefile.bkup

@@ -1,96 +0,0 @@
-#-----------------------------------------------------------------------------#
-
-PARQUETDIR := /home/mario/Desktop/Record_Evolution/parquet-cpp
-ARROWDIR := /home/mario/Desktop/Record_Evolution/arrow/cpp/src
-
-CPP := g++ -std=c++14
-OPT := -Wall -Woverflow -Wpedantic -Wextra -Waddress -Waligned-new -Walloc-zero
-
-prepare : collect_parquet modify_parquet collect_arrow modify_arrow
-
-collect_parquet :
-	cp -r $(PARQUETDIR)/src/parquet ./
-	cp $(PARQUETDIR)/examples/low-level-api/reader_writer.h ./
-	cp $(PARQUETDIR)/examples/low-level-api/reader-writer.cc ./
-
-modify_parquet :
-	cp parquet/parquet_version.h.in parquet/parquet_version.h
-	sed -i 's/ReadableFileInterface/ReadWriteFileInterface/g' parquet/util/memory.h
-	sed -i 's/ReadableFileInterface/ReadWriteFileInterface/g' parquet/file_reader.h
-	sed -i 's/arrow::Codec/arrow::util::Codec/g' parquet/util/memory.h
-	sed -i 's/valid_bits_writer/valid_bits_offset/g' parquet/column_reader.h
-
-collect_arrow :
-	cp -r $(ARROWDIR)/arrow ./
-
-modify_arrow :
-	cp arrow/util/bit_util.h arrow/util/bit-util.h
-
-collect_test :
-	cp $(PARQUETDIR)/examples/low-level-api/reader-writer.cc ./
-
-subst :
-	sed -i 's/#include \"arrow\//\/\/#include \"arrow/g' parquet/properties.h
-
-test :
-	$(CPP) $(OPT) -I$(PWD) reader-writer.cc
-
-clean :
-	rm -r parquet/ arrow/
-	rm reader-writer.cc reader_writer.h
-
-#-----------------------------------------------------------------------------#
-
-# choose shell
-SHELL:=/bin/bash
-
-SRC = reader-writer
-
-# specify path of cloned directory
-ARROWGIT := /home/mario/Desktop/Record_Evolution/arrow
-
-filewriter : parquet/file_writer.cc
-		$(CPP) -c $(OPT) $<
-
-# build executable (and generate dependency file)
-readwrite : reader-writer.cc
-		$(CPP) $(OPT) -MMD $< -I ./
-
-# generate dependency file
-$(SRC).d : $(SRC).cc
-		$(CPP) -c -MMD $< -I ./ -I $(ARROWGIT)/cpp/src/
-
-# extract source dependencies
-extract-dep : $(SRC).d
-		@# extract relevant dependencies
-		cat $< | sed 's/ /\n/g' | awk 'NF' | grep -v '\\' | grep '\/' > deps.log
-		cat deps.log | sed ':a;N;$!ba;s/\n/ /g' > headers.log
-		cat headers.log | sed 's/.h$$/.cc/g' > sources.log
-		@# copy required sources
-		mkdir -p temp/
-		cp --parents `cat headers.log` temp/
-		cp --parents `cat sources.log` temp/ 2>/dev/null
-		mv temp$(ARROWGIT)/cpp/src/* ./
-		rm -r temp
-
-clean-dep :
-		rm -f deps.log headers.log sources.log $(SRC).d
-
-#-----------------------------------------------------------------------------#
-# only use more recent and up to date repository arrow.git
-
-# build arrow shared/static libraries
-build :
-	cd arrow/cpp
-	# cmake -LA  to show all options
-	cmake . -D ARROW_PARQUET=ON #ARROW_ARMV8_ARCH=armv8-a
-	make
-
-example :
-	cd arrow/cpp/examples/parquet/low-level-api/
-	g++ reader-writer.cc -I. -I../../../src/ -L../../../../cpp/build/release/ -larrow -lparquet
-
-# set environment variable LD_LIBRARY_PATH=../../../../cpp/build/release/ before launching executable
-
-
-#------------------------------------------------------------------------------------#

parquet/parquet/makefileobj

@@ -1,303 +0,0 @@
-bin/type.o : src/src/arrow/type.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/result.o : src/src/arrow/result.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder.o : src/src/arrow/builder.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/tensor.o : src/src/arrow/tensor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/table.o : src/src/arrow/table.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/extension_type.o : src/src/arrow/extension_type.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/device.o : src/src/arrow/device.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/memory_pool.o : src/src/arrow/memory_pool.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/datum.o : src/src/arrow/datum.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/record_batch.o : src/src/arrow/record_batch.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/compare.o : src/src/arrow/compare.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/visitor.o : src/src/arrow/visitor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/chunked_array.o : src/src/arrow/chunked_array.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/status.o : src/src/arrow/status.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/pretty_print.o : src/src/arrow/pretty_print.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/sparse_tensor.o : src/src/arrow/sparse_tensor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/buffer.o : src/src/arrow/buffer.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/scalar.o : src/src/arrow/scalar.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/string.o : src/src/arrow/util/string.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/memory.o : src/src/arrow/util/memory.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/future.o : src/src/arrow/util/future.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/iterator.o : src/src/arrow/util/iterator.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/compression.o : src/src/arrow/util/compression.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/utf8.o : src/src/arrow/util/utf8.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/time.o : src/src/arrow/util/time.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/cpu_info.o : src/src/arrow/util/cpu_info.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/thread_pool.o : src/src/arrow/util/thread_pool.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bit_util.o : src/src/arrow/util/bit_util.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/logging.o : src/src/arrow/util/logging.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/basic_decimal.o : src/src/arrow/util/basic_decimal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/decimal.o : src/src/arrow/util/decimal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bit_block_counter.o : src/src/arrow/util/bit_block_counter.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/key_value_metadata.o : src/src/arrow/util/key_value_metadata.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/int_util.o : src/src/arrow/util/int_util.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/io_util.o : src/src/arrow/util/io_util.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bitmap_ops.o : src/src/arrow/util/bitmap_ops.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bitmap_builders.o : src/src/arrow/util/bitmap_builders.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bit_run_reader.o : src/src/arrow/util/bit_run_reader.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/value_parsing.o : src/src/arrow/util/value_parsing.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/string_builder.o : src/src/arrow/util/string_builder.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/formatting.o : src/src/arrow/util/formatting.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_primitive.o : src/src/arrow/array/array_primitive.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_dict.o : src/src/arrow/array/array_dict.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_binary.o : src/src/arrow/array/builder_binary.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_union.o : src/src/arrow/array/builder_union.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/concatenate.o : src/src/arrow/array/concatenate.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_nested.o : src/src/arrow/array/array_nested.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_decimal.o : src/src/arrow/array/array_decimal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_primitive.o : src/src/arrow/array/builder_primitive.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/data.o : src/src/arrow/array/data.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/diff.o : src/src/arrow/array/diff.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_nested.o : src/src/arrow/array/builder_nested.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_decimal.o : src/src/arrow/array/builder_decimal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_dict.o : src/src/arrow/array/builder_dict.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_binary.o : src/src/arrow/array/array_binary.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_adaptive.o : src/src/arrow/array/builder_adaptive.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/array_base.o : src/src/arrow/array/array_base.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/validate.o : src/src/arrow/array/validate.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/builder_base.o : src/src/arrow/array/builder_base.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/util.o : src/src/arrow/array/util.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/caching.o : src/src/arrow/io/caching.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/memory.o : src/src/arrow/io/memory.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/interfaces.o : src/src/arrow/io/interfaces.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/buffered.o : src/src/arrow/io/buffered.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/file.o : src/src/arrow/io/file.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/strtod.o : src/src/arrow/vendored/double-conversion/strtod.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bignum.o : src/src/arrow/vendored/double-conversion/bignum.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/fixed-dtoa.o : src/src/arrow/vendored/double-conversion/fixed-dtoa.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/fast-dtoa.o : src/src/arrow/vendored/double-conversion/fast-dtoa.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/diy-fp.o : src/src/arrow/vendored/double-conversion/diy-fp.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/double-conversion.o : src/src/arrow/vendored/double-conversion/double-conversion.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/bignum-dtoa.o : src/src/arrow/vendored/double-conversion/bignum-dtoa.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/cached-powers.o : src/src/arrow/vendored/double-conversion/cached-powers.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/api_aggregate.o : src/src/arrow/compute/api_aggregate.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/exec.o : src/src/arrow/compute/exec.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/kernel.o : src/src/arrow/compute/kernel.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/registry.o : src/src/arrow/compute/registry.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/function.o : src/src/arrow/compute/function.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/cast.o : src/src/arrow/compute/cast.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/api_vector.o : src/src/arrow/compute/api_vector.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/api_scalar.o : src/src/arrow/compute/api_scalar.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/codegen_internal.o : src/src/arrow/compute/kernels/codegen_internal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/column_scanner.o : src/src/parquet/column_scanner.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/statistics.o : src/src/parquet/statistics.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/internal_file_decryptor.o : src/src/parquet/internal_file_decryptor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/column_writer.o : src/src/parquet/column_writer.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/encryption.o : src/src/parquet/encryption.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/file_reader.o : src/src/parquet/file_reader.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/properties.o : src/src/parquet/properties.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/encryption_internal.o : src/src/parquet/encryption_internal.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/internal_file_encryptor.o : src/src/parquet/internal_file_encryptor.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/types.o : src/src/parquet/types.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/encoding.o : src/src/parquet/encoding.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/metadata.o : src/src/parquet/metadata.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/printer.o : src/src/parquet/printer.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/level_conversion.o : src/src/parquet/level_conversion.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/deprecated_io.o : src/src/parquet/deprecated_io.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/file_writer.o : src/src/parquet/file_writer.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/schema.o : src/src/parquet/schema.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/platform.o : src/src/parquet/platform.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-
-bin/column_reader.o : src/src/parquet/column_reader.cc
-	$(CPP) $(CPPFLAGS) -c $< $(LIBS) -o $@
-

parquet/parquet/reader-writer.cc

@@ -1,413 +0,0 @@
-// Licensed to the Apache Software Foundation (ASF) under one
-// or more contributor license agreements. See the NOTICE file
-// distributed with this work for additional information
-// regarding copyright ownership. The ASF licenses this file
-// to you under the Apache License, Version 2.0 (the
-// "License"); you may not use this file except in compliance
-// with the License. You may obtain a copy of the License at
-//
-// http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing,
-// software distributed under the License is distributed on an
-// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
-// KIND, either express or implied. See the License for the
-// specific language governing permissions and limitations
-// under the License.
-
-#include <cassert>
-#include <fstream>
-#include <iostream>
-#include <memory>
-
-#include "reader_writer.h"
-
-/*
- * This example describes writing and reading Parquet Files in C++ and serves as a
- * reference to the API.
- * The file contains all the physical data types supported by Parquet.
- * This example uses the RowGroupWriter API that supports writing RowGroups optimized for
- *memory consumption
- **/
-
-/* Parquet is a structured columnar file format
- * Parquet File = "Parquet data" + "Parquet Metadata"
- * "Parquet data" is simply a vector of RowGroups. Each RowGroup is a batch of rows in a
- * columnar layout
- * "Parquet Metadata" contains the "file schema" and attributes of the RowGroups and their
- * Columns
- * "file schema" is a tree where each node is either a primitive type (leaf nodes) or a
- * complex (nested) type (internal nodes)
- * For specific details, please refer the format here:
- * https://github.com/apache/parquet-format/blob/master/LogicalTypes.md
- **/
-
-constexpr int NUM_ROWS_PER_ROW_GROUP = 500;
-const char PARQUET_FILENAME[] = "parquet_cpp_example.parquet";
-
-int main(int argc, char** argv) {
-  /**********************************************************************************
-                             PARQUET WRITER EXAMPLE
-  **********************************************************************************/
-  // parquet::REQUIRED fields do not need definition and repetition level values
-  // parquet::OPTIONAL fields require only definition level values
-  // parquet::REPEATED fields require both definition and repetition level values
-  try {
-    // Create a local file output stream instance.
-    using FileClass = ::arrow::io::FileOutputStream;
-    std::shared_ptr<FileClass> out_file;
-    PARQUET_ASSIGN_OR_THROW(out_file, FileClass::Open(PARQUET_FILENAME));
-
-    // Setup the parquet schema
-    std::shared_ptr<GroupNode> schema = SetupSchema();
-
-    // Add writer properties
-    parquet::WriterProperties::Builder builder;
-    builder.compression(parquet::Compression::UNCOMPRESSED);
-    std::shared_ptr<parquet::WriterProperties> props = builder.build();
-
-    // Create a ParquetFileWriter instance
-    std::shared_ptr<parquet::ParquetFileWriter> file_writer =
-        parquet::ParquetFileWriter::Open(out_file, schema, props);
-
-    // Append a RowGroup with a specific number of rows.
-    parquet::RowGroupWriter* rg_writer = file_writer->AppendRowGroup();
-
-    // Write the Bool column
-    parquet::BoolWriter* bool_writer =
-        static_cast<parquet::BoolWriter*>(rg_writer->NextColumn());
-    for (int i = 0; i < NUM_ROWS_PER_ROW_GROUP; i++) {
-      bool value = ((i % 2) == 0) ? true : false;
-      bool_writer->WriteBatch(1, nullptr, nullptr, &value);
-    }
-
-    // Write the Int32 column
-    parquet::Int32Writer* int32_writer =
-        static_cast<parquet::Int32Writer*>(rg_writer->NextColumn());
-    for (int i = 0; i < NUM_ROWS_PER_ROW_GROUP; i++) {
-      int32_t value = i;
-      int32_writer->WriteBatch(1, nullptr, nullptr, &value);
-    }
-
-    // Write the Int64 column. Each row has repeats twice.
-    parquet::Int64Writer* int64_writer =
-        static_cast<parquet::Int64Writer*>(rg_writer->NextColumn());
-    for (int i = 0; i < 2 * NUM_ROWS_PER_ROW_GROUP; i++) {
-      int64_t value = i * 1000 * 1000;
-      value *= 1000 * 1000;
-      int16_t definition_level = 1;
-      int16_t repetition_level = 0;
-      if ((i % 2) == 0) {
-        repetition_level = 1;  // start of a new record
-      }
-      int64_writer->WriteBatch(1, &definition_level, &repetition_level, &value);
-    }
-
-    // Write the INT96 column.
-    parquet::Int96Writer* int96_writer =
-        static_cast<parquet::Int96Writer*>(rg_writer->NextColumn());
-    for (int i = 0; i < NUM_ROWS_PER_ROW_GROUP; i++) {
-      parquet::Int96 value;
-      value.value[0] = i;
-      value.value[1] = i + 1;
-      value.value[2] = i + 2;
-      int96_writer->WriteBatch(1, nullptr, nullptr, &value);
-    }
-
-    // Write the Float column
-    parquet::FloatWriter* float_writer =
-        static_cast<parquet::FloatWriter*>(rg_writer->NextColumn());
-    for (int i = 0; i < NUM_ROWS_PER_ROW_GROUP; i++) {
-      float value = static_cast<float>(i) * 1.1f;
-      float_writer->WriteBatch(1, nullptr, nullptr, &value);
-    }
-
-    // Write the Double column
-    parquet::DoubleWriter* double_writer =
-        static_cast<parquet::DoubleWriter*>(rg_writer->NextColumn());
-    for (int i = 0; i < NUM_ROWS_PER_ROW_GROUP; i++) {
-      double value = i * 1.1111111;
-      double_writer->WriteBatch(1, nullptr, nullptr, &value);
-    }
-
-    // Write the ByteArray column. Make every alternate values NULL
-    parquet::ByteArrayWriter* ba_writer =
-        static_cast<parquet::ByteArrayWriter*>(rg_writer->NextColumn());
-    for (int i = 0; i < NUM_ROWS_PER_ROW_GROUP; i++) {
-      parquet::ByteArray value;
-      char hello[FIXED_LENGTH] = "parquet";
-      hello[7] = static_cast<char>(static_cast<int>('0') + i / 100);
-      hello[8] = static_cast<char>(static_cast<int>('0') + (i / 10) % 10);
-      hello[9] = static_cast<char>(static_cast<int>('0') + i % 10);
-      if (i % 2 == 0) {
-        int16_t definition_level = 1;
-        value.ptr = reinterpret_cast<const uint8_t*>(&hello[0]);
-        value.len = FIXED_LENGTH;
-        ba_writer->WriteBatch(1, &definition_level, nullptr, &value);
-      } else {
-        int16_t definition_level = 0;
-        ba_writer->WriteBatch(1, &definition_level, nullptr, nullptr);
-      }
-    }
-
-    // Write the FixedLengthByteArray column
-    parquet::FixedLenByteArrayWriter* flba_writer =
-        static_cast<parquet::FixedLenByteArrayWriter*>(rg_writer->NextColumn());
-    for (int i = 0; i < NUM_ROWS_PER_ROW_GROUP; i++) {
-      parquet::FixedLenByteArray value;
-      char v = static_cast<char>(i);
-      char flba[FIXED_LENGTH] = {v, v, v, v, v, v, v, v, v, v};
-      value.ptr = reinterpret_cast<const uint8_t*>(&flba[0]);
-
-      flba_writer->WriteBatch(1, nullptr, nullptr, &value);
-    }
-
-    // Close the ParquetFileWriter
-    file_writer->Close();
-
-    // Write the bytes to file
-    DCHECK(out_file->Close().ok());
-  } catch (const std::exception& e) {
-    std::cerr << "Parquet write error: " << e.what() << std::endl;
-    return -1;
-  }
-
-  /**********************************************************************************
-                             PARQUET READER EXAMPLE
-  **********************************************************************************/
-
-  try {
-    // Create a ParquetReader instance
-    std::unique_ptr<parquet::ParquetFileReader> parquet_reader =
-        parquet::ParquetFileReader::OpenFile(PARQUET_FILENAME, false);
-
-    // Get the File MetaData
-    std::shared_ptr<parquet::FileMetaData> file_metadata = parquet_reader->metadata();
-
-    // Get the number of RowGroups
-    int num_row_groups = file_metadata->num_row_groups();
-    assert(num_row_groups == 1);
-
-    // Get the number of Columns
-    int num_columns = file_metadata->num_columns();
-    assert(num_columns == 8);
-
-    // Iterate over all the RowGroups in the file
-    for (int r = 0; r < num_row_groups; ++r) {
-      // Get the RowGroup Reader
-      std::shared_ptr<parquet::RowGroupReader> row_group_reader =
-          parquet_reader->RowGroup(r);
-
-      int64_t values_read = 0;
-      int64_t rows_read = 0;
-      int16_t definition_level;
-      int16_t repetition_level;
-      int i;
-      std::shared_ptr<parquet::ColumnReader> column_reader;
-
-      ARROW_UNUSED(rows_read); // prevent warning in release build
-
-      // Get the Column Reader for the boolean column
-      column_reader = row_group_reader->Column(0);
-      parquet::BoolReader* bool_reader =
-          static_cast<parquet::BoolReader*>(column_reader.get());
-
-      // Read all the rows in the column
-      i = 0;
-      while (bool_reader->HasNext()) {
-        bool value;
-        // Read one value at a time. The number of rows read is returned. values_read
-        // contains the number of non-null rows
-        rows_read = bool_reader->ReadBatch(1, nullptr, nullptr, &value, &values_read);
-        // Ensure only one value is read
-        assert(rows_read == 1);
-        // There are no NULL values in the rows written
-        assert(values_read == 1);
-        // Verify the value written
-        bool expected_value = ((i % 2) == 0) ? true : false;
-        assert(value == expected_value);
-        i++;
-      }
-
-      // Get the Column Reader for the Int32 column
-      column_reader = row_group_reader->Column(1);
-      parquet::Int32Reader* int32_reader =
-          static_cast<parquet::Int32Reader*>(column_reader.get());
-      // Read all the rows in the column
-      i = 0;
-      while (int32_reader->HasNext()) {
-        int32_t value;
-        // Read one value at a time. The number of rows read is returned. values_read
-        // contains the number of non-null rows
-        rows_read = int32_reader->ReadBatch(1, nullptr, nullptr, &value, &values_read);
-        // Ensure only one value is read
-        assert(rows_read == 1);
-        // There are no NULL values in the rows written
-        assert(values_read == 1);
-        // Verify the value written
-        assert(value == i);
-        i++;
-      }
-
-      // Get the Column Reader for the Int64 column
-      column_reader = row_group_reader->Column(2);
-      parquet::Int64Reader* int64_reader =
-          static_cast<parquet::Int64Reader*>(column_reader.get());
-      // Read all the rows in the column
-      i = 0;
-      while (int64_reader->HasNext()) {
-        int64_t value;
-        // Read one value at a time. The number of rows read is returned. values_read
-        // contains the number of non-null rows
-        rows_read = int64_reader->ReadBatch(1, &definition_level, &repetition_level,
-                                            &value, &values_read);
-        // Ensure only one value is read
-        assert(rows_read == 1);
-        // There are no NULL values in the rows written
-        assert(values_read == 1);
-        // Verify the value written
-        int64_t expected_value = i * 1000 * 1000;
-        expected_value *= 1000 * 1000;
-        assert(value == expected_value);
-        if ((i % 2) == 0) {
-          assert(repetition_level == 1);
-        } else {
-          assert(repetition_level == 0);
-        }
-        i++;
-      }
-
-      // Get the Column Reader for the Int96 column
-      column_reader = row_group_reader->Column(3);
-      parquet::Int96Reader* int96_reader =
-          static_cast<parquet::Int96Reader*>(column_reader.get());
-      // Read all the rows in the column
-      i = 0;
-      while (int96_reader->HasNext()) {
-        parquet::Int96 value;
-        // Read one value at a time. The number of rows read is returned. values_read
-        // contains the number of non-null rows
-        rows_read = int96_reader->ReadBatch(1, nullptr, nullptr, &value, &values_read);
-        // Ensure only one value is read
-        assert(rows_read == 1);
-        // There are no NULL values in the rows written
-        assert(values_read == 1);
-        // Verify the value written
-        parquet::Int96 expected_value;
-        ARROW_UNUSED(expected_value); // prevent warning in release build
-        expected_value.value[0] = i;
-        expected_value.value[1] = i + 1;
-        expected_value.value[2] = i + 2;
-        for (int j = 0; j < 3; j++) {
-          assert(value.value[j] == expected_value.value[j]);
-        }
-        i++;
-      }
-
-      // Get the Column Reader for the Float column
-      column_reader = row_group_reader->Column(4);
-      parquet::FloatReader* float_reader =
-          static_cast<parquet::FloatReader*>(column_reader.get());
-      // Read all the rows in the column
-      i = 0;
-      while (float_reader->HasNext()) {
-        float value;
-        // Read one value at a time. The number of rows read is returned. values_read
-        // contains the number of non-null rows
-        rows_read = float_reader->ReadBatch(1, nullptr, nullptr, &value, &values_read);
-        // Ensure only one value is read
-        assert(rows_read == 1);
-        // There are no NULL values in the rows written
-        assert(values_read == 1);
-        // Verify the value written
-        float expected_value = static_cast<float>(i) * 1.1f;
-        assert(value == expected_value);
-        i++;
-      }
-
-      // Get the Column Reader for the Double column
-      column_reader = row_group_reader->Column(5);
-      parquet::DoubleReader* double_reader =
-          static_cast<parquet::DoubleReader*>(column_reader.get());
-      // Read all the rows in the column
-      i = 0;
-      while (double_reader->HasNext()) {
-        double value;
-        // Read one value at a time. The number of rows read is returned. values_read
-        // contains the number of non-null rows
-        rows_read = double_reader->ReadBatch(1, nullptr, nullptr, &value, &values_read);
-        // Ensure only one value is read
-        assert(rows_read == 1);
-        // There are no NULL values in the rows written
-        assert(values_read == 1);
-        // Verify the value written
-        double expected_value = i * 1.1111111;
-        assert(value == expected_value);
-        i++;
-      }
-
-      // Get the Column Reader for the ByteArray column
-      column_reader = row_group_reader->Column(6);
-      parquet::ByteArrayReader* ba_reader =
-          static_cast<parquet::ByteArrayReader*>(column_reader.get());
-      // Read all the rows in the column
-      i = 0;
-      while (ba_reader->HasNext()) {
-        parquet::ByteArray value;
-        // Read one value at a time. The number of rows read is returned. values_read
-        // contains the number of non-null rows
-        rows_read =
-            ba_reader->ReadBatch(1, &definition_level, nullptr, &value, &values_read);
-        // Ensure only one value is read
-        assert(rows_read == 1);
-        // Verify the value written
-        char expected_value[FIXED_LENGTH] = "parquet";
-        ARROW_UNUSED(expected_value); // prevent warning in release build
-        expected_value[7] = static_cast<char>('0' + i / 100);
-        expected_value[8] = static_cast<char>('0' + (i / 10) % 10);
-        expected_value[9] = static_cast<char>('0' + i % 10);
-        if (i % 2 == 0) {  // only alternate values exist
-          // There are no NULL values in the rows written
-          assert(values_read == 1);
-          assert(value.len == FIXED_LENGTH);
-          assert(memcmp(value.ptr, &expected_value[0], FIXED_LENGTH) == 0);
-          assert(definition_level == 1);
-        } else {
-          // There are NULL values in the rows written
-          assert(values_read == 0);
-          assert(definition_level == 0);
-        }
-        i++;
-      }
-
-      // Get the Column Reader for the FixedLengthByteArray column
-      column_reader = row_group_reader->Column(7);
-      parquet::FixedLenByteArrayReader* flba_reader =
-          static_cast<parquet::FixedLenByteArrayReader*>(column_reader.get());
-      // Read all the rows in the column
-      i = 0;
-      while (flba_reader->HasNext()) {
-        parquet::FixedLenByteArray value;
-        // Read one value at a time. The number of rows read is returned. values_read
-        // contains the number of non-null rows
-        rows_read = flba_reader->ReadBatch(1, nullptr, nullptr, &value, &values_read);
-        // Ensure only one value is read
-        assert(rows_read == 1);
-        // There are no NULL values in the rows written
-        assert(values_read == 1);
-        // Verify the value written
-        char v = static_cast<char>(i);
-        char expected_value[FIXED_LENGTH] = {v, v, v, v, v, v, v, v, v, v};
-        assert(memcmp(value.ptr, &expected_value[0], FIXED_LENGTH) == 0);
-        i++;
-      }
-    }
-  } catch (const std::exception& e) {
-    std::cerr << "Parquet read error: " << e.what() << std::endl;
-    return -1;
-  }
-
-  std::cout << "Parquet Writing and Reading Complete" << std::endl;
-
-  return 0;
-}

parquet/parquet/reader_writer.h

@@ -1,71 +0,0 @@
-// Licensed to the Apache Software Foundation (ASF) under one
-// or more contributor license agreements. See the NOTICE file
-// distributed with this work for additional information
-// regarding copyright ownership. The ASF licenses this file
-// to you under the Apache License, Version 2.0 (the
-// "License"); you may not use this file except in compliance
-// with the License. You may obtain a copy of the License at
-//
-// http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing,
-// software distributed under the License is distributed on an
-// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
-// KIND, either express or implied. See the License for the
-// specific language governing permissions and limitations
-// under the License.
-
-#include <arrow/io/file.h>
-#include <arrow/util/logging.h>
-
-#include <parquet/api/reader.h>
-#include <parquet/api/writer.h>
-
-using parquet::ConvertedType;
-using parquet::Repetition;
-using parquet::Type;
-using parquet::schema::GroupNode;
-using parquet::schema::PrimitiveNode;
-
-constexpr int FIXED_LENGTH = 10;
-
-static std::shared_ptr<GroupNode> SetupSchema() {
-  parquet::schema::NodeVector fields;
-  // Create a primitive node named 'boolean_field' with type:BOOLEAN,
-  // repetition:REQUIRED
-  fields.push_back(PrimitiveNode::Make("boolean_field", Repetition::REQUIRED,
-                                       Type::BOOLEAN, ConvertedType::NONE));
-
-  // Create a primitive node named 'int32_field' with type:INT32, repetition:REQUIRED,
-  // logical type:TIME_MILLIS
-  fields.push_back(PrimitiveNode::Make("int32_field", Repetition::REQUIRED, Type::INT32,
-                                       ConvertedType::TIME_MILLIS));
-
-  // Create a primitive node named 'int64_field' with type:INT64, repetition:REPEATED
-  fields.push_back(PrimitiveNode::Make("int64_field", Repetition::REPEATED, Type::INT64,
-                                       ConvertedType::NONE));
-
-  fields.push_back(PrimitiveNode::Make("int96_field", Repetition::REQUIRED, Type::INT96,
-                                       ConvertedType::NONE));
-
-  fields.push_back(PrimitiveNode::Make("float_field", Repetition::REQUIRED, Type::FLOAT,
-                                       ConvertedType::NONE));
-
-  fields.push_back(PrimitiveNode::Make("double_field", Repetition::REQUIRED, Type::DOUBLE,
-                                       ConvertedType::NONE));
-
-  // Create a primitive node named 'ba_field' with type:BYTE_ARRAY, repetition:OPTIONAL
-  fields.push_back(PrimitiveNode::Make("ba_field", Repetition::OPTIONAL, Type::BYTE_ARRAY,
-                                       ConvertedType::NONE));
-
-  // Create a primitive node named 'flba_field' with type:FIXED_LEN_BYTE_ARRAY,
-  // repetition:REQUIRED, field_length = FIXED_LENGTH
-  fields.push_back(PrimitiveNode::Make("flba_field", Repetition::REQUIRED,
-                                       Type::FIXED_LEN_BYTE_ARRAY, ConvertedType::NONE,
-                                       FIXED_LENGTH));
-
-  // Create a GroupNode named 'schema' using the primitive nodes defined above
-  // This GroupNode is the root node of the schema tree
-  return std::static_pointer_cast<GroupNode>(
-      GroupNode::Make("schema", Repetition::REQUIRED, fields));
-}

parquet/parquet/src_copy.sh

@@ -1,6 +0,0 @@
-#!/bin/bash
-
-mkdir src
-cat deps.log | while read f; do cp --parents $f src/; done;
-mv src/home/mario/Desktop/arrow/cpp/* src/
-rm -r src/home/

parquet/parquet/src_setup.sh

@@ -1,171 +0,0 @@
-#!/bin/bash
-
-#-----------------------------------------------------------------------------#
-
-# NOTE: before starting to extract the minimal required sources and dependencies
-#       run
-#       $ cd cpp/
-#       $ cmake -D ARROW_PARQUET=ON
-#      in the arrow repository
-
-# provide
-# - local path of clone of https://github.com/apache/arrow.git
-# - name/path of main .hpp file of cython extension
-repo="$1"
-main="$2"
-depf="$3"
-
-# check CLI arguments
-if [ -z "$repo" ] || [ -z "$main" ] || [ -z "$depf" ]; then
-  echo "please provide..."
-  echo "1. local path of arrow repository"
-  echo "2. name of main .hpp/.cpp"
-  echo "3. desired name of dependency file"
-  echo -e "example:\n./setup-sources.sh /home/mario/Desktop/Record_Evolution/arrow/ reader-writer.cc deps.log"
-  exit 1
-fi
-
-echo -e "extracting sources from/for \n1: ${repo}\n2: ${main}\n3: ${depf}\n"
-
-# make sure the dependency file is empty
-rm -f ${depf}
-touch ${depf}
-
-# define maximal recursion depth
-maxdep=8
-
-#-----------------------------------------------------------------------------#
-
-# define function to list dependencies of source file in repository recursively
-listDependencies()
-{
-  rep="$1"
-  src="$2"
-  dep="$3"
-  rec="$4"
-
-  echo -e "\nstarting 'listDependencies()' for\n1. ${rep}\n2. ${src}\n3. ${dep}\n4. ${rec}"
-
-  # generate dependency file (and remove resulting object file)
-  echo -e "g++ -c -MMD ${src} -I ${rep}cpp/src/\n"
-  g++ -c -MMD ${src} -I ${rep}cpp/src/
-
-  # derive name of dependency and object files
-  depf=$(basename ${src} | sed 's/.cc/.d/g')
-  objf=$(basename ${src} | sed 's/.cc/.o/g')
-  rm ${objf}
-
-  # list dependencies by
-  # 1. removing header
-  # 2. remove source itself
-  # 3. delete leading spaces
-  # 4. delete trailing backslashs
-  # 5. remove empty lines
-  cat ${depf} | grep ${rep} | grep -v ${src} | tr -d "^ " | tr -d "\\\\" | awk 'NF' > listdep.log
-  # rm ${depf}
-
-  while IFS= read -r fs
-  do
-    echo "$fs"
-    # check if dependency is already in the list
-    if grep -Fxq "$fs" "$dep"
-    then
-      echo "dep exist"
-    else
-      echo "dep does not exist yet => adding it"
-      # add dependency to list
-      echo "$fs" >> ${dep}
-      # check for corresponding source file
-      fssourc=$(echo ${fs} | sed 's/.h$/.cc/g' | sed 's/.hpp$/.cpp/g')
-      echo ${fssourc}
-      if [ -f "$fssourc" ]
-      then
-        echo "source file exists"
-        # list nested dependencies
-        if [ "$rec" -lt "$maxdep" ]
-        then
-          # increment recursion depth
-          recinc=$(($rec+1))
-          # call recursion
-          listDependencies ${rep} ${fssourc} ${dep} ${recinc}
-        else
-          echo "maximal recursion depth exceeded"
-        fi
-      else
-        echo "source file does not exist"
-      fi
-    fi
-    echo ""
-  done < listdep.log
-
-  # cat listdep.log | while read fs
-  # do
-  #   echo $fs
-  #   # check if dependency is already in the list
-  #   inlist=$(cat listdep.log | grep ${fs} | wc -l)
-  #   echo ${inlist}
-  #   # check for any corresponding source files
-  #   # if [ -f ]
-  # done
-}
-
-#-----------------------------------------------------------------------------#
-
-# call function to list dependencies (recursively)
-listDependencies ${repo} ${main} ${depf} 0
-
-# # generate dependency file (and remove resulting object file)
-# echo -e "generate dependencies:\ng++ -c -MMD ${main} -I ./ -I ${repo}cpp/src/\n"
-# g++ -c -MMD ${main} -I ${repo}cpp/src/
-# rm $(echo ${main} | sed 's/.cc/.o/g')
-#
-# # derive name of dependency file
-# dep=$(echo ${main} | sed 's/.cc/.d/g')
-#
-# if [ -f "$dep" ]; then
-#
-#   # list dependencies
-#   cat ${dep} | sed 's/ /\n/g' | awk 'NF' | grep -v '\\' | grep '\/' > deps.log
-#
-#   # extract list of headers
-#   cat deps.log | sed ':a;N;$!ba;s/\n/ /g' > deps-headers.log
-#   echo "list of required headers ($(cat deps.log | wc -l))"
-#   cat deps-headers.log
-#   echo ""
-#
-#   # imply list of sources
-#   cat deps.log | sed 's/.h$/.cc/g' | sed 's/.hpp$/.cpp/g' > sources_raw.log
-#   cat sources_raw.log | while read f
-#   do
-#     if [ -f "$f" ]; then
-#       echo $f >> sources_check.log
-#     fi
-#   done
-#   cat sources_check.log | sed ':a;N;$!ba;s/\n/ /g' > deps-sources.log
-#   echo "list of required sources ($(cat sources_check.log | wc -l))"
-#   cat deps-sources.log
-#   echo ""
-#
-#   # remove all temporary files
-#   rm ${dep} deps.log
-#   rm sources_raw.log sources_check.log
-#
-#   # copy required headers and sources
-#   echo -e "copy required headers and sources"
-#   mkdir temp/
-#   cp --parents `cat deps-headers.log` temp/
-#   cp --parents `cat deps-sources.log` temp/
-#   mv temp${repo}cpp/src/* ./
-#   rm -r temp
-#
-#   # remove dependencies
-#   #rm deps-headers.log deps-sources.log
-#
-#   # show files
-#   ls -lh
-#
-# else
-#
-#   echo -e "\nERROR: failed to generate dependency file\n"
-#
-# fi

parquet/pyarrow_arm/Dockerfile

@@ -1,26 +0,0 @@
-
-FROM ubuntu:19.10
-
-RUN apt-get update -y && apt-get install -y \
-	apt-utils \
-	git g++ \
-	make cmake \
-	pkg-config \
-	#build-essentials \
-	python3 \
-	python3-setuptools \
-	cython3 \
-	python3-numpy
-
-RUN git clone https://github.com/apache/arrow.git --single-branch --depth=1
-
-COPY . ./
-
-RUN chmod u+x ./build_arrow_cpp.sh
-RUN chmod u+x ./build_arrow_python.sh
-RUN ./build_arrow_cpp.sh
-RUN ./build_arrow_python.sh
-
-#RUN chmod u+x ./build_arrow.sh
-#CMD ["./build_arrow.sh"]
-CMD ["sleep 1d"]

parquet/pyarrow_arm/build_and_test.sh

@@ -1,5 +0,0 @@
-#!/bin/bash
-
-docker build . --tag=pyarrowbuild:latest
-
-docker run -it pyarrowbuild:latest /bin/bash

parquet/pyarrow_arm/build_arrow.sh

@@ -1,65 +0,0 @@
-#!/bin/bash
-
-sleep infinity
-
-startts=$(date)
-echo "starting build process at ${startts}..."
-
-echo -e "\nhome directory is..."
-pwd
-
-echo -e "\ncloning apache/arrow..."
-git clone https://github.com/apache/arrow.git --single-branch --depth=1
-
-echo -e "\nls -lh /\n"
-ls -lh /
-
-echo -e "\nls -lh arrow/\n"
-ls -lh arrow/
-
-echo -e "\nls -lh arrow/python/\n"
-ls -lh arrow/python
-
-
-mkdir arrow/cpp/build
-pushd arrow/cpp/build
-
-cmake -DCMAKE_INSTALL_PREFIX=$ARROW_HOME \
-      -DCMAKE_INSTALL_LIBDIR=lib \
-      -DARROW_WITH_BZ2=ON \
-      -DARROW_WITH_ZLIB=ON \
-      -DARROW_WITH_ZSTD=ON \
-      -DARROW_WITH_LZ4=ON \
-      -DARROW_WITH_SNAPPY=ON \
-      -DARROW_WITH_BROTLI=ON \
-      -DARROW_PARQUET=ON \
-      -DARROW_PYTHON=ON \
-      -DARROW_BUILD_TESTS=OFF \
-      -DARROW_WITH_HDFS=OFF \
-      ..
-
-make -j4
-make install
-popd
-
-#cython --version
-cython3 --version
-
-pushd arrow/python
-export ARROW_LIB_DIR=/lib/
-export PYARROW_WITH_PARQUET=1
-export PYARROW_WITH_CUDA=0
-export PYARROW_WITH_FlIGHT=0
-export PYARROW_WITH_DATASET=0
-export PYARROW_WITH_ORC=0
-export PYARROW_WITH_PLASMA=0
-export PYARROW_WITH_S3FS=0
-export PYARROW_WITH_HDFS=0
-export PYARROW_WITH_GANDIVA=0
-python3 setup.py build_ext --inplace
-popd
-
-echo "  started build process at ${startts} ..."
-finishts=$(date)
-echo "finishing build process at ${finishts}..."
-

parquet/pyarrow_arm/build_arrow_cpp.sh

@@ -1,23 +0,0 @@
-#!/bin/bash
-
-mkdir arrow/cpp/build
-pushd arrow/cpp/build
-
-cmake -DCMAKE_INSTALL_PREFIX=$ARROW_HOME \
-      -DCMAKE_INSTALL_LIBDIR=lib \
-      -DARROW_WITH_BZ2=ON \
-      -DARROW_WITH_ZLIB=ON \
-      -DARROW_WITH_ZSTD=ON \
-      -DARROW_WITH_LZ4=ON \
-      -DARROW_WITH_SNAPPY=ON \
-      -DARROW_WITH_BROTLI=ON \
-      -DARROW_PARQUET=ON \
-      -DARROW_PYTHON=ON \
-      -DARROW_BUILD_TESTS=OFF \
-      -DARROW_WITH_HDFS=OFF \
-      -DARROW_WITH_IPC=OFF \
-      ..
-
-make -j4
-make install
-popd

parquet/pyarrow_arm/build_arrow_python.sh

@@ -1,15 +0,0 @@
-#!/bin/bash
-
-pushd arrow/python
-export PYARROW_WITH_PARQUET=1
-export PYARROW_WITH_CUDA=0
-export PYARROW_WITH_FlIGHT=0
-export PYARROW_WITH_DATASET=0
-export PYARROW_WITH_ORC=0
-export PYARROW_WITH_PLASMA=0
-export PYARROW_WITH_S3FS=0
-export PYARROW_WITH_HDFS=0
-export PYARROW_WITH_GANDIVA=0
-# python3 setup.py build_ext --inplace
-python3 setup.py install
-popd

parquet/pyarrow_arm/makefile

@@ -1,23 +0,0 @@
-
-build :
-	docker build . --tag pyarrowbuild
-
-run :
-	docker run -it pyarrowbuild:latest
-
-run-bash :
-	docker run -it --volume=$(pwd)/build:/home pyarrowbuild:latest /bin/bash
-
-run-volume :
-	docker run -it -v /home/pirate/pyarrow/build/:/arrow/python/ pyarrowbuild:latest
-	#sudo docker run -it --volume=$(pwd)/build:/home ubuntu:latest /bin/bash
-
-rm-container :
-	cont=$(docker ps -a | tail -n 26 | awk '{print $NF}' | sed ':a;N;$!ba;s/\n/ /g')
-	echo ${cont}
-	docker rm ${cont}
-
-rm-image :
-	img=$(docker image ls --quiet | sed ':a;N;$!ba;s/\n/ /g')
-	docker image rm ${img}
-

parquet/pyarrow_arm/pyarrow_test.py

@@ -1,18 +0,0 @@
-
-import pyarrow.parquet as pq
-import pyarrow.csv as pv
-
-csvfile = 'pressureVacuum.csv'
-
-tb = pv.read_csv(csvfile,parse_options=pv.ParseOptions(delimiter=','))
-
-print(tb)
-
-parquetfile = 'pressureVacuum.parquet'
-
-pq.write_table(tb,parquetfile,compression='BROTLI')
-#  {'NONE', 'SNAPPY', 'GZIP', 'LZO', 'BROTLI', 'LZ4', 'ZSTD'}
-
-df = pq.read_table(parquetfile,columns=None)
-
-print(df)

parquet/pyarrow_arm/sync_pi.sh

@@ -1,8 +0,0 @@
-#!/bin/bash
-
-if [ -z "$1" ]
-then
-	exit 1
-fi
-
-scp $1 pirate@mf-pi-40:/home/pirate/pyarrow/

pip/bkup.pyproject.toml

@@ -1,4 +0,0 @@
-[build-system]
-requires = [
-    "setuptools"
-]

pip/makefile

@@ -1,29 +0,0 @@
-# --------------------------------------------------------------------------- #
-
-SHELL := /bin/bash
-
-publish: sdist upload
-
-sdist: ../cython/py_imc_termite.pyx ../cython/imc_termite.pxd ../cython/py_imc_termite.cpp
-	cp -v $? ./
-	cp -v $(shell ls ../lib/imc_*.hpp) ./
-	tail -n 212 ../README.md > ./README.md
-	cp -v ../LICENSE ./
-	python3 setup.py sdist
-
-# authentication:
-# - username: __token__
-# - password: <token value including pypi-prefix>
-upload:
-	python3 -m twine upload dist/$(shell ls -t dist/ | head -n1)
-
-clean:
-	rm -rvf dist/
-	rm -rvf *.egg-info
-	rm -rvf build/
-	rm -rvf cython/
-	rm -vf *.pyx *.pxd
-	rm -vf *.cpp *.c *.hpp
-	rm -vf README.md LICENSE
-
-# --------------------------------------------------------------------------- #

pip/setup.py

@@ -1,36 +0,0 @@
-
-from setuptools import setup, Extension
-
-with open("README.md", "r", encoding="utf-8") as fh:
-    long_description = fh.read()
-
-setup(
-    name="IMCtermite",
-    version="1.2.5",
-    author="Record Evolution GmbH",
-    author_email="mario.fink@record-evolution.de",
-    maintainer="Record Evolution GmbH",
-    license="MIT",
-    description="Enables extraction of measurement data from binary files with extension 'raw' used by proprietary software imcFAMOS/imcSTUDIO and facilitates its storage in open source file formats",
-    keywords="IMC raw imcFAMOS imcSTUDIO imcCRONOS",
-    long_description=long_description,
-    long_description_content_type="text/markdown",
-    url="https://github.com/RecordEvolution/IMCtermite.git",
-    project_urls={
-        "Bug Tracker": "https://github.com/RecordEvolution/IMCtermite/issues",
-    },
-    classifiers=[
-        "Programming Language :: Python :: 3",
-        "License :: OSI Approved :: MIT License",
-        "Operating System :: OS Independent",
-    ],
-    ext_modules=[Extension("imc_termite",
-                           ["py_imc_termite.cpp"],
-                           # libraries_dirs=["cython/"],
-                           # include_dirs=["3rdparty/pugixml/","lib/"],
-                           # depends=["../lib/tdm_termite.hpp"]
-                           language='c++',
-                           extra_compile_args=['-std=c++17','-Wno-unused-variable'],
-                           extra_link_args=['-std=c++17'],
-                           )],
-)

python/MANIFEST.in

@@ -1,4 +1,5 @@
-include *.hpp
+include lib/*.hpp
 include *.cpp
 include *.pyx
 include *.pxd
+include VERSION

python/VERSION

@@ -0,0 +1 @@
+2.1.18

python/example.py

@@ -1,137 +0,0 @@
-
-#-----------------------------------------------------------------------------#
-
-import raw_eater
-import raw_meat
-import pyarrow as pa
-import pyarrow.parquet as pq
-from pathlib import Path
-
-fileobj1 = Path("samples/datasetA/").rglob("*.raw")
-rawlist1 = [str(fl) for fl in fileobj1]
-
-fileobj2 = Path("samples/datasetB/").rglob("*.raw")
-rawlist2 = [str(fl) for fl in fileobj2]
-
-rawlist = rawlist1 #[rawlist1[0],rawlist1[4],rawlist2[0],rawlist2[6]]
-for fil in rawlist2 :
-    rawlist.append(fil)
-rawlist.append("./README.md")
-
-print("")
-print(rawlist)
-print()
-
-#-----------------------------------------------------------------------------#
-
-# alternatively create "empty" instance of "raw_eater" and set file names
-eatraw = raw_eater.raweater()
-# eatraw.set_file("../smp/pressure_Vacuum.raw".encode())
-
-# convert every single listed file
-for rf in rawlist :
-
-    print("converting " + str(rf) + "...\n" + 90*("-") + "\n")
-
-    # setup instance of "raw_eater" and trigger conversion
-    # eatraw = raw_eater.raweater(rf.encode())
-    # eatraw = raw_meat.rawmerger(rf.encode())
-
-    # use global instance of "raw_eater" to set file and perform decoding
-    eatraw.set_file(rf.encode())
-    try :
-        eatraw.do_conversion()
-    except RuntimeError as e :
-        print("conversion failed: " + str(e))
-
-    # check validity of file format
-    if eatraw.validity() :
-
-        # show channel name and its unit
-        entity = eatraw.channel_name().decode(encoding='UTF-8',errors='ignore')
-        unit = eatraw.unit().decode(encoding='UTF-8',errors='ignore')
-        print("\nentity: " + str(entity))
-        print("unit:   " + str(unit) + "\n")
-
-        # obtain extracted data
-        xt = eatraw.get_time()
-        yt = eatraw.get_channel()
-
-        # show excerpt of data
-        print("time (length: " + str(len(xt)) + ") \n"
-             + str(xt[:10]) + "\n...\n" + str(xt[-10:]) + "\n")
-        yttrunc = [round(y,4) for y in yt]
-        print(str(entity) + " (length: " + str(len(yttrunc)) + ") \n"
-            + str(yttrunc[:10]) + "\n...\n" + str(yttrunc[-10:]) + "\n")
-
-        outname = rf.split('/')[-1].replace('raw','csv')
-
-        print("write output to : " + outname)
-        eatraw.write_table(("output/"+outname).encode(),ord(' '))
-
-    else :
-
-        print("\nerror: invalid/corrupt .raw file")
-
-    print("\n")
-
-#-----------------------------------------------------------------------------#
-
-print("convert and merge channels " + "\n" + 90*("-") + "\n")
-
-# setup new instance to merge channels
-eatmea = raw_meat.rawmerger(''.encode()) #rawlist[0].encode())
-
-# add every single channel/file in list
-for rf in rawlist :
-    print("\nadding channel " + str(rf))
-    try :
-        succ = eatmea.add_channel(rf.encode())
-        print("\nrecent time series: length: " + str(len(eatmea.get_time_series())) + "\n")
-    except RuntimeError as e :
-        print("failed to add channel: " + str(e))
-
-# show summary of successfully merged channels
-print("\nmerged channels:\n")
-
-# write merged table to .csv output
-eatmea.write_table_all('output/allchannels.csv'.encode(),ord(','))
-
-# get number of successfully merged channels and their names (+units)
-numch = eatmea.get_num_channels()
-chnames = [chnm.decode(encoding='UTF-8',errors='ignore') for chnm in eatmea.get_channel_names()]
-print("number of channels: " + str(numch))
-print("channel names: " + str(chnames))
-
-# obtain final time series
-timse = eatmea.get_time_series()
-print("\nfinal time series:\nlength:" + str(len(timse)) + "\n")
-
-# get time unit and prepend column name
-chnames.insert(0,"Time ["+str(eatmea.time_unit().decode(encoding='UTF-8',errors='ignore'))+"]")
-
-# prepare list of pyarrow arrays
-pyarrs = []
-pyarrs.append(pa.array(timse))
-
-for i in range(0,numch) :
-    print("\n" + str(i) + "  " + str(chnames[i]))
-    dat = eatmea.get_channel_by_index(i)
-    print("length: " + str(len(dat)))
-    pyarrs.append(pa.array(dat))
-print("")
-# print("\npyarrow arrays\n" + str(pyarrs))
-
-# create pyarrow table from data
-pyarwtab = pa.Table.from_arrays(pyarrs,chnames)
-print("\n" + 60*"-" + "\n" + str(pyarwtab) + "\n")
-
-# write pyarrow table to .parquet file with compression
-pq.write_table(pyarwtab,'output/allchannels.parquet',compression='BROTLI')  # compression='BROTLI', 'SNAPPY')
-
-# try to read and decode the .parquet file
-df = pq.read_table('output/allchannels.parquet')
-print(df.to_pandas())
-# df.to_pandas().to_csv('allchannels.csv',index=False,encoding='utf-8',sep=",")
-
-#-----------------------------------------------------------------------------#

python/examples/multichannel.py

@@ -0,0 +1,43 @@
+
+import imctermite
+import pandas
+import datetime
+
+def add_trigger_time(trigger_time, add_time) :
+    trgts = datetime.datetime.strptime(trigger_time,'%Y-%m-%dT%H:%M:%S')
+    dt = datetime.timedelta(seconds=add_time)
+    return (trgts + dt).strftime('%Y-%m-%dT%H:%M:%S:%f')
+
+if __name__ == "__main__" :
+
+    # read file and extract data
+    imctm = imctermite.imctermite(b"Measurement.raw")
+    chns = imctm.get_channels(True)
+    
+    # prepare abscissa
+    xcol = "time ["+chns[0]['xunit']+"]"
+    #xcol = "timestamp"
+    xsts = [add_trigger_time(chns[0]['trigger-time'],tm) for tm in chns[0]['xdata']]
+
+    # sort channels
+    chnnms = sorted([chn['name'] for chn in chns], reverse=False)
+    chnsdict = {}
+    for chn in chns :
+        chnsdict[chn['name']] = chn
+
+    # construct dataframe
+    df = pandas.DataFrame()
+    df[xcol] = pandas.Series(chns[0]['xdata'])
+    #df[xcol] = pandas.Series(xsts)
+    #for idx,chn in enumerate(chns) :
+    for chnnm in chnnms :
+        chn = chnsdict[chnnm]
+        #xcol = (chn['xname'] if chn['xname'] != '' else "x_"+str(idx))+" ["+chn['xunit']+"]"
+        #df[xcol] = pandas.Series(chn['xdata'])
+        ycol = chn['yname']+" ["+chn['yunit']+"]"
+        df[ycol] = pandas.Series(chn['ydata'])
+
+    # show entire dataframe and write file
+    print(df)
+    df.to_csv("Measurement.csv",header=True,sep='\t',index=False)
+

python/examples/usage.py

@@ -1,11 +1,11 @@
 
-import imc_termite
+import imctermite
 import json
 import os
 
 # declare and initialize instance of "imctermite" by passing a raw-file
 try :
-    imcraw = imc_termite.imctermite(b"samples/exampleB.raw")
+    imcraw = imctermite.imctermite(b"samples/exampleB.raw")
 except RuntimeError as e :
     raise Exception("failed to load/parse raw-file: " + str(e))
 
@@ -21,18 +21,18 @@ if len(channelsdata) > 0 :
 
 print(len(chnydata))
 print(len(chnxdata))
+print()
 
 # print the channels into a specific directory
-imcraw.print_channels(b"./data",ord(','))
+imcraw.print_channels(b"/tmp/",ord(','))
 
 # print all channels separately
-idx = 0
-for chn in channels :
-    print(str(idx)+" : "+chn['name']+" : "+chn['uuid'])
-    filname = os.path.join("./data",str(idx) + "_" + chn['name']+".csv")
+for i,chn in enumerate(channels) :
+    print(str(i)+" : "+chn['name']+" : "+chn['uuid'])
+    filname = os.path.join("/tmp/",str(i) + "_" + chn['name']+".csv")
     print(filname)
     imcraw.print_channel(chn['uuid'].encode(),filname.encode(),ord(','))
-    idx = idx + 1
 
 # print all channels in single file
-# imcraw.print_table(b"./data/allchannels.csv")
+imcraw.print_table(b"/tmp/allchannels.csv")
+

python/examples/usage_adv.py

@@ -1,5 +1,5 @@
 
-import imc_termite
+import imctermite
 import json
 import os
 
@@ -15,7 +15,7 @@ for fl in rawlist1:
 
     # declare and initialize instance of "imctermite" by passing a raw-file
     try :
-        imcraw = imc_termite.imctermite(fl.encode())
+        imcraw = imctermite.imctermite(fl.encode())
     except RuntimeError as e :
         raise Exception("failed to load/parse raw-file: " + str(e))
 
@@ -24,7 +24,7 @@ for fl in rawlist1:
     print(json.dumps(channels,indent=4, sort_keys=False))
 
     # print the channels into a specific directory
-    imcraw.print_channels(b"./")
+    imcraw.print_channels(b"./",ord(','))
 
     # print all channels in single file
     imcraw.print_table(("./"+str(os.path.basename(fl).split('.')[0])+"_allchannels.csv").encode())

python/examples/usage_ext.py

@@ -0,0 +1,50 @@
+
+import imctermite
+import json
+import os
+import datetime
+
+# declare and initialize instance of "imctermite" by passing a raw-file
+try :
+    imcraw = imctermite.imctermite(b"samples/sampleB.raw")
+except RuntimeError as e :
+    raise Exception("failed to load/parse raw-file: " + str(e))
+
+# obtain list of channels as list of dictionaries (without data)
+channels = imcraw.get_channels(False)
+print(json.dumps(channels,indent=4, sort_keys=False))
+
+# obtain all channels (including full data)
+channelsdata = imcraw.get_channels(True)
+
+# everything that follows is an example that specifically makes use only of
+# the first (index = 0) channel ...
+idx = 0
+
+if len(channelsdata) > 0 :
+
+    # get first channel's data
+    chnydata = channelsdata[idx]['ydata']
+    chnxdata = channelsdata[idx]['xdata']
+    print("xdata: " + str(len(chnxdata)))
+    print("ydata: " + str(len(chnydata)))
+
+    # extract trigger-time
+    trigtim = datetime.datetime.fromisoformat(channels[idx]["trigger-time"])
+    print(trigtim)
+
+    # file output of data with absolute timestamp in 1st column
+    filname = os.path.join("./",channelsdata[idx]['name']+".csv")
+    print("writing output into " + filname)
+    with open(filname,'w') as fout :
+        # include column header
+        fout.write( str(channelsdata[idx]['xname']) + '[' + str(channelsdata[idx]['xunit']) + "]"
+                  + ","
+                  + str(channelsdata[idx]['yname']) + '[' + str(channelsdata[idx]['yunit']) + "]"
+                  + "\n" )
+        # add data (introduce time shift according to trigger-time)
+        for row in range(0,len(chnxdata)) :
+            fout.write( str( (trigtim + datetime.timedelta(seconds=chnxdata[row])).isoformat() )
+                      + ","
+                      + str( chnydata[row])
+                      + "\n" )

python/examples/usage_files.py

@@ -0,0 +1,29 @@
+
+import imctermite import imctermite
+
+def show_results(imcraw) :
+
+    channels = imcraw.get_channels(False)
+    print(channels)
+
+    channelsData = imcraw.get_channels(True)
+    print("number of channels: " + str(len(channelsData)))
+
+    for (i,chn) in enumerate(channelsData) :
+        print(str(i) +  " | " + chn['name'])
+        print(chn['xname'] + " | " + chn['xunit'])
+        print(chn['xdata'][:10])
+        print(chn['yname'] + " | " + chn['yunit'])
+        print(chn['ydata'][:10])
+
+    print("")
+
+# create instance of 'imctermite'
+imcraw = imctermite(b'samples/sampleA.raw')
+
+show_results(imcraw)
+
+# use previous instance of 'imctermite' to provide new file
+imcraw.submit_file(b'samples/sampleB.raw')
+
+show_results(imcraw)

python/imctermite.pxd

@@ -1,19 +1,23 @@
-# cython: language_level = 3
 
 # use some C++ STL libraries
 from libcpp.string cimport string
 from libcpp.vector cimport vector
 from libcpp cimport bool
 
-cdef extern from "imc_raw.hpp" namespace "imc":
-  cdef cppclass imc_termite "imc::raw":
+cdef extern from "lib/imc_raw.hpp" namespace "imc":
+
+  cdef cppclass cppimctermite "imc::raw":
+
     # constructor(s)
-    imc_termite() except +
-    imc_termite(string rawfile) except +
+    cppimctermite() except +
+    cppimctermite(string rawfile) except +
+
     # provide raw file
     void set_file(string rawfile) except +
+
     # get JSON list of channels
     vector[string] get_channels(bool json, bool data) except +
+
     # print single channel/all channels
     void print_channel(string channeluuid, string outputdir, char delimiter) except +
     void print_channels(string outputdir, char delimiter) except +

python/imctermite.pyx

@@ -1,39 +1,49 @@
 # distutils: language = c++
+# cython: language_level = 3
 
-from imc_termite cimport imc_termite
+from imctermite cimport cppimctermite
 
 import json as jn
 import decimal
-# import numpy as np
+import platform
+
+# auxiliary function for codepage conversion
+def get_codepage(chn) :
+    if platform == 'Windows' :
+        chndec = jn.loads(chn.decode(errors="ignore"))
+        chncdp = chndec["codepage"]
+        return 'utf-8' if chncdp is None else chncdp
+    else :
+        return 'utf-8'
 
 cdef class imctermite:
 
   # C++ instance of class => stack allocated (requires nullary constructor!)
-  cdef imc_termite cpp_imc
+  cdef cppimctermite cppimc
 
   # constructor
   def __cinit__(self, string rawfile):
-    self.cpp_imc = imc_termite(rawfile)
+    self.cppimc = cppimctermite(rawfile)
 
   # provide raw file
   def submit_file(self,string rawfile):
-    self.cpp_imc.set_file(rawfile)
+    self.cppimc.set_file(rawfile)
 
   # get JSON list of channels
-  def get_channels(self, bool data):
-    chnlst = self.cpp_imc.get_channels(True,data)
-    chnlstjn = [jn.loads(chn.decode(errors="ignore")) for chn in chnlst]
+  def get_channels(self, bool include_data):
+    chnlst = self.cppimc.get_channels(True,include_data)
+    chnlstjn = [jn.loads(chn.decode(get_codepage(chn),errors="ignore")) for chn in chnlst]
     return chnlstjn
 
   # print single channel/all channels
   def print_channel(self, string channeluuid, string outputfile, char delimiter):
-    self.cpp_imc.print_channel(channeluuid,outputfile,delimiter)
+    self.cppimc.print_channel(channeluuid,outputfile,delimiter)
   def print_channels(self, string outputdir, char delimiter):
-    self.cpp_imc.print_channels(outputdir,delimiter)
+    self.cppimc.print_channels(outputdir,delimiter)
 
   # print table including channels
   def print_table(self, string outputfile):
-    chnlst = self.cpp_imc.get_channels(True,True)
+    chnlst = self.cppimc.get_channels(True,True)
     chnlstjn = [jn.loads(chn.decode(errors="ignore")) for chn in chnlst]
     with open(outputfile.decode(),'w') as fout:
       for chn in chnlstjn:

python/makefile

@@ -0,0 +1,46 @@
+
+setup:
+	cat ../README.md | grep '^# IMCtermite' -A 50000 > ./README.md
+	#pandoc -f markdown -t rst -o README.rst README.md
+	#python -m rstvalidator README.rst
+	cp -r ../lib ./
+	cp -v ../LICENSE ./
+
+setup-clean:
+	rm -vf README.md README.rst LICENSE
+	rm -rf lib/
+
+build: setup
+	python setup.py build
+
+build-inplace: setup
+	python setup.py build_ext --inplace
+
+build-sdist: setup
+	python setup.py sdist
+	python -m twine check dist/*
+
+build-bdist: setup
+	python setup.py bdist
+	python -m twine check dist/*
+
+build-clean:
+	python setup.py clean --all
+	rm -vf imctermite*.so imctermite*.cpp
+	rm -vf IMCtermite*.so IMCtermite*.cpp
+	rm -rvf dist/ IMCtermite.egg-info/
+	rm -rvf dist/ imctermite.egg-info/
+
+cibuildwheel-build: setup
+	cibuildwheel --platform linux
+
+cibuildwheel-clean:
+	rm -rvf wheelhouse/
+
+pypi-upload:
+	python -m twine upload dist/$(shell ls -t dist/ | head -n1)
+
+clean: setup build-clean cibuildwheel-clean setup-clean
+
+run-example:
+	PYTHONPATH=$(pwd) python examples/usage_files.py

python/numpy_to_pyarrow.py

@@ -1,24 +0,0 @@
-
-import pyarrow as pa
-import numpy as np
-import pyarrow.parquet as pq
-
-db = pa.array(np.linspace(10,50,6))
-print(db)
-da = pa.array(np.linspace(0,5,6))
-print(db)
-
-filenam = 'pyarrow_testtab.parquet'
-
-patab = pa.Table.from_arrays([da,db],['entity A [unitA]','entity B [unitB]'])
-print(patab)
-
-# pq.write_table(patab,filenam,compression='BROTLI')
-pq.write_table(patab,filenam,compression='SNAPPY')
-
-df = pq.read_table(filenam)
-print(df)
-print(df.to_pandas())
-
-#import readline
-#readline.write_history_file('generate_pyarrow_table_and_write_parquet.py')

python/pyproject.toml

@@ -0,0 +1,6 @@
+[build-system]
+requires = ["setuptools", "wheel","Cython"]
+build-backend = "setuptools.build_meta"
+
+[tool.cibuildwheel]
+before-all = ""

python/setup.cfg

@@ -0,0 +1,23 @@
+
+[metadata]
+name = imctermite
+description = Enables extraction of measurement data from binary files with extension 'raw' used by proprietary software imcFAMOS and imcSTUDIO and facilitates its storage in open source file formats
+long_description = file: README.md
+# long_description_content_type = text/x-rst
+long_description_content_type = text/markdown
+version = file: VERSION
+author = Record Evolution GmbH
+author_email = mario.fink@record-evolution.de
+maintainer = Record Evolution GmbH
+url= https://github.com/RecordEvolution/IMCtermite.git
+license = MIT License
+license_files = LICENSE
+keywords = IMC, raw, imcFAMOS, imcSTUDIO, imcCRONOS
+classifiers =
+  Programming Language :: Python :: 3
+  License :: OSI Approved :: MIT License
+  Operating System :: OS Independent
+  Topic :: Scientific/Engineering
+  Topic :: Software Development :: Libraries :: Python Modules
+
+[options]

python/setup.py

@@ -0,0 +1,21 @@
+from setuptools import Extension, setup
+from Cython.Build import cythonize
+import sys
+
+print("building on platform: "+sys.platform)
+
+cmpArgs = {
+    "linux": ['-std=c++17','-Wno-unused-variable'],
+    "darwin": ['-std=c++17','-Wno-unused-variable'],
+    "win32": ['/EHsc','/std:c++17']
+}
+
+extension = Extension(
+    "imctermite",
+    sources=["imctermite.pyx"],
+    extra_compile_args=cmpArgs[sys.platform]
+)
+
+setup(
+    ext_modules=cythonize(extension,language_level=3)
+)

samples/exampleA-20230124.raw

@@ -0,0 +1,14 @@
+|CF,2,1,1;|CK,1,3,1,1;
+|Nv,1,12,7,3,4,64,1,0;
+|NO,1,12,1,5,FAMOS,0,;
+|NL,1,10,1252,0x407;
+|CG,1,5,1,1,1;
+|CD,1,13,1,1,1,s,0,0,0;
+|NT,1,19, 6, 4,2018,11,33,54;
+|CC,1,3,1,1;
+|CP,1,16,1,8,8,64,0,0,1,0;
+|Cb,1,22,1,0,1,1,0,8,0,8,1,0,0,;
+|CR,1,11,0,0,0,1,1,V;
+|CN,1,35,0,0,0,7,Average,16,Measurement 0815;
+|CT,1,43,0,8,TxTester,8,E. Smith,16,Measurement 0815;
+|CS,1,10,1,ÍÌÌÌÌÌ(@;

samples/exampleB-20230124.raw

@@ -0,0 +1,24 @@
+|CF,2,1,1;|CK,1,3,1,1;
+|Nv,1,12,7,3,4,64,1,0;
+|NO,1,12,1,5,FAMOS,0,;
+|NL,1,10,1252,0x407;
+|CB,1,12,1,5,Meas1,0,;
+|CG,1,5,1,1,1;
+|CD,1,16,5E-1,1,1,2,0,0,0;
+|NT,1,19, 6, 4,2018,11,24,18;
+|CC,1,3,1,1;
+|CP,1,15,1,1,2,8,0,0,1,0;
+|Cb,1,22,1,0,1,1,0,3,0,3,1,3,0,;
+|CR,1,30,1,3.937007874015748E-2,5,1,1,V;
+|ND,1,15,-1,-1,-1,0,1E+1;
+|CN,1,16,1,0,0,5,Chan1,0,;
+|CG,1,5,1,1,1;
+|CD,1,16,5E-1,1,1,2,0,0,0;
+|NT,1,19, 6, 4,2018,11,24,18;
+|CC,1,3,1,1;
+|CP,1,15,2,1,2,8,0,0,1,0;
+|Cb,1,22,1,0,2,1,3,3,0,3,1,3,0,;
+|CR,1,30,1,3.937007874015748E-2,5,1,1,V;
+|ND,1,15,-1,-1,-1,0,1E+1;
+|CN,1,16,1,0,0,5,Chan2,0,;
+|CS,1,8,1, šš šš;

samples/exampleB.raw

@@ -8,7 +8,7 @@
 |CR,1,49,1,3.921568627450980E-2,0.000000000000000E+0,1,1,V;
 |ND,1,50,-1,-1,-1,0.000000000000000E+0,1.000000000000000E+1;|CN,1,17,1,0,0,6,kanal2,0,;
 |CS,1,8,1,leclass="Code Example" translate="true">
-|NO,1,37,1,9,                   imc-FAMOS,20,Erzeugt:E.Mustermann; |CG,1,5,2,2,2;
+|NO,1,37,1,9,imc-FAMOS,20,Erzeugt:E.Mustermann; |CG,1,5,2,2,2;
 |CD,1,31,1.000000000000000E-1,1,0,,0,0,0;|NT,1,27,10, 6,1993,19,18,20.0000000;
 |CC,1,3,1,1;|CP,1,16,1,4,7,32,0,0,1,0;|Cb,1,40,1,0,1,1,0,16,0,16,1,0.000000000000E+0,0,;
 |CR,1,11,0,0,0,1,1,V;|CN,1,20,0,0,0,9,E06_6_121,0,;|CC,1,3,2,1;|CP,1,16,2,4,5,32,0,0,1,0;

samples/exampleC-20230124.raw

@@ -0,0 +1,17 @@
+|CF,2,1,1;|CK,1,3,1,1;
+|Nv,1,12,7,3,4,64,1,0;
+|NO,1,12,1,5,FAMOS,0,;
+|NL,1,10,1252,0x407;
+|CG,1,5,2,2,2;
+|CD,1,12,1,1,0,,2,0,0;
+|NT,1,18, 6, 4,2018,11,37,1;
+|CC,1,3,1,1;
+|CP,1,16,1,4,7,32,0,0,1,0;
+|Cb,1,24,1,0,1,1,0,16,0,16,1,0,0,;
+|CR,1,11,0,0,0,1,1,V;
+|CN,1,20,0,0,0,9,MyXY_plot,0,;
+|CC,1,3,2,1;
+|CP,1,16,2,2,3,16,0,0,1,0;
+|Cb,1,23,1,0,2,1,16,8,0,8,1,0,0,;
+|CR,1,30,1,4.577706569008927E-5,0,1,1,s;
+|CS,1,26,1, @ €? @ffF@ UUªªÿÿ;

shell/check_markers.sh

@@ -1,12 +1,50 @@
 ##!/bin/bash/
 
-dir=$1
+fildir=$1
 
-#ls ${dir} | while read fn; do echo $fn; cat ${dir}$fn | grep -a "|[A-Z][A-Z]," -o | wc -l; done;
+if [ -z "${fildir}" ]; then
+  echo "CLI argument missing: provide file or directory" >&2
+  exit 1
+fi
 
-#ls ${dir} | while read fn; do echo $fn; cat ${dir}$fn | grep -a "|[A-Z][A-Z]," -o; done;
+list-markers()
+{
+  arg="$1"
+  if [ -z "${arg}" ]; then
+    echo "list-markers: missing file argument" >&2
+    exit 1
+  else
+    if [ -d "${arg}" ]; then
+      echo "list-markers: file argument is a directory" >&2
+      exit 1
+    fi
+  fi
   
-#ls ${dir} | while read fn; do echo $fn; cat ${dir}$fn | xxd | head -n10 | tail -n3; done;
+  echo -e "\n${arg}\n"
+  mrks=$(cat ${arg} | grep -a "|[A-Z][a-zA-Z]," -o)
+  mrksnum=$(echo "${mrks}" | wc -l)
+  echo -e "(${mrksnum})\n${mrks}"
+}
+
+if [ -f "${fildir}" ]; then
+
+  echo "analyzing single file ${fildir} ..."
+  list-markers "${fildir}"
+
+elif [ -d "${fildir}" ]; then
+
+  echo "analyzing all *.raw files in directory ${fildir} ..."
+  lsfls=$(ls ${fildir}/*.raw | sed 's/\/\//\//g')
+  echo -e "\n${lsfls}"
+
+  for fl in ${lsfls}; do
+    list-markers "${fl}"
+  done
+
+else
+
+  echo "does not exist: ${fildir}" >&2
+  exit 1
+
+fi
 
-ls ${dir} | while read fn; do echo $fn; cat ${dir}$fn | grep -a "|[A-Z][a-zA-Z]," -o | wc -l; done;
-ls ${dir} | while read fn; do echo $fn; cat ${dir}$fn | grep -a "|[A-Z][a-zA-Z]," -o; done;

src/main.cpp

@@ -12,6 +12,7 @@
 
 const std::string gittag("TAGSTRING");
 const std::string githash("HASHSTRING");
+const std::string timestamp("TIMESTAMPSTRING");
 
 //---------------------------------------------------------------------------//
 
@@ -127,13 +128,13 @@ optkeys parse_args(int argc, char* argv[], bool list_args = false)
 
 void show_version()
 {
-  std::cout<<"imctermite ["<<gittag<<"-g"<<githash<<"]"<<"\n";
+  std::cout<<"imctermite ["<<gittag<<"-g"<<githash<<"-"<<timestamp<<"]"<<"\n";
 }
 
 void show_usage()
 {
   std::cout<<"\n"
-           <<"imctermite ["<<gittag<<"-g"<<githash<<"] (github.com/RecordEvolution/IMCtermite.git)"
+           <<"imctermite ["<<gittag<<"-g"<<githash<<"-"<<timestamp<<"] (https://github.com/RecordEvolution/IMCtermite.git)"
            <<"\n\n"
            <<"Decode IMC raw files and dump data as *.csv"
            <<"\n\n"