Mark required arguments in scripts as required

.gitlab-ci.yml

@@ -1,8 +1,5 @@
 image: docker:19.03.13
 
-variables:
-  DOCKER_TLS_CERTDIR: "/certs"
-
 services:
   - docker:19.03.13-dind
 
@@ -10,6 +7,10 @@ stages:
   - build
   - push
 
+variables:
+  DOCKER_TLS_CERTDIR: "/certs"
+  INTERMEDIATE_IMAGE_TAG: $CI_REGISTRY_IMAGE:$CI_COMMIT_REF_NAME-$CI_COMMIT_SHA
+
 .reg_setup:
   before_script:
     - apk add --no-cache curl
@@ -28,8 +29,6 @@ build_image:
   stage: build
   tags:
     - docker
-  variables:
-    INTERMEDIATE_IMAGE_TAG: $CI_REGISTRY_IMAGE:$CI_COMMIT_SHA
 
 push_master:
   extends:
@@ -47,7 +46,6 @@ push_master:
     - docker
   variables:
     IMAGE_TAG: $CI_REGISTRY_IMAGE:latest
-    INTERMEDIATE_IMAGE_TAG: $CI_REGISTRY_IMAGE:$CI_COMMIT_SHA
 
 push_other:
   extends:
@@ -68,4 +66,3 @@ push_other:
     - docker
   variables:
     IMAGE_TAG: $CI_REGISTRY_IMAGE:$CI_COMMIT_REF_NAME
-    INTERMEDIATE_IMAGE_TAG: $CI_REGISTRY_IMAGE:$CI_COMMIT_SHA

Dockerfile

@@ -7,28 +7,36 @@ LABEL authors="Patrick Jentsch <p.jentsch@uni-bielefeld.de>, Stephan Porada <por
 ENV LANG=C.UTF-8
 
 
-RUN apt-get update
+RUN apt-get update \
+ && apt-get install --no-install-recommends --yes \
+      procps \
+      python3.7 \
+      python3-pip \
+      wget \
+ && python3 -m pip install \
+      chardet \
+      setuptools \
+      wheel
 
-
+# Install the NLP pipeline and it's dependencies #
-# Install pipeline dependencies #
 ## Install pyFlow ##
-ENV PYFLOW_RELEASE=1.1.20
+ENV PYFLOW_VERSION=1.1.20
-ADD "https://github.com/Illumina/pyflow/releases/download/v${PYFLOW_RELEASE}/pyflow-${PYFLOW_RELEASE}.tar.gz" .
+RUN wget --no-check-certificate --quiet \
-RUN tar -xzf "pyflow-${PYFLOW_RELEASE}.tar.gz" \
+      "https://github.com/Illumina/pyflow/releases/download/v${PYFLOW_VERSION}/pyflow-${PYFLOW_VERSION}.tar.gz" \
- && cd "pyflow-${PYFLOW_RELEASE}" \
+ && tar -xzf "pyflow-${PYFLOW_VERSION}.tar.gz" \
+ && cd "pyflow-${PYFLOW_VERSION}" \
  && apt-get install --no-install-recommends --yes \
       python2.7 \
  && python2.7 setup.py build install \
- && cd .. \
+ && cd - > /dev/null \
- && rm -r "pyflow-${PYFLOW_RELEASE}" "pyflow-${PYFLOW_RELEASE}.tar.gz"
+ && rm -r "pyflow-${PYFLOW_VERSION}" "pyflow-${PYFLOW_VERSION}.tar.gz"
 
 
 ## Install spaCy ##
-ENV SPACY_VERSION=3.0.3
+ENV SPACY_VERSION=3.2.1
 RUN apt-get install --no-install-recommends --yes \
       python3.7 \
       python3-pip \
-      zip \
  && pip3 install \
       chardet \
       setuptools \
@@ -36,22 +44,20 @@ RUN apt-get install --no-install-recommends --yes \
  && pip3 install --upgrade pip \
  && pip3 install "spacy==${SPACY_VERSION}"
 
-ENV SPACY_MODELS_VERSION=3.0.0
+
-RUN python3 -m spacy download "da_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ENV SPACY_MODELS="de_core_news_md,en_core_web_md,it_core_news_md,pl_core_news_md,zh_core_web_md"
- && python3 -m spacy download "de_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ENV SPACY_MODELS_VERSION=3.2.0
- && python3 -m spacy download "el_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+RUN for spacy_model in $(echo ${SPACY_MODELS} | tr "," "\n"); do python3 -m spacy download "${spacy_model}-${SPACY_MODELS_VERSION}" --direct; done
- && python3 -m spacy download "en_core_web_md-${SPACY_MODELS_VERSION}" --direct \
+
- && python3 -m spacy download "es_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+
- && python3 -m spacy download "fr_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+COPY packages .
- && python3 -m spacy download "it_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+RUN cd stand-off-data-py \
- && python3 -m spacy download "nl_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ && python3 -m pip install . \
- && python3 -m spacy download "pt_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ && cd -
- && python3 -m spacy download "ru_core_news_md-${SPACY_MODELS_VERSION}" --direct \
- && python3 -m spacy download "zh_core_web_md-${SPACY_MODELS_VERSION}" --direct
 
 
 ## Install Pipeline ##
-COPY nlp spacy-nlp /usr/local/bin/
+COPY nlp spacy-nlp vrt-creator /usr/local/bin/
 
 
 RUN rm -r /var/lib/apt/lists/*

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2021 Bielefeld University - CRC 1288 - INF
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -1,88 +1,41 @@
 # NLP - Natural Language Processing
 
-This software implements a heavily parallelized pipeline for Natural Language Processing of text files. It is used for nopaque's NLP service but you can also use it standalone, for that purpose a convenient wrapper script is provided.
+This software implements a heavily parallelized pipeline for Natural Language Processing of text files. It is used for nopaque's NLP service but you can also use it standalone, for that purpose a convenient wrapper script is provided. The pipeline is designed to run on Linux operating systems, but with some tweaks it should also run on Windows with WSL installed.
 
 ## Software used in this pipeline implementation
-- Official Debian Docker image (buster-slim) and programs from its free repositories: https://hub.docker.com/_/debian
+
+- Official Debian Docker image (buster-slim): https://hub.docker.com/_/debian
+  - Software from Debian Buster's free repositories
 - pyFlow (1.1.20): https://github.com/Illumina/pyflow/releases/tag/v1.1.20
-- spaCy (3.0.3): https://github.com/tesseract-ocr/tesseract/releases/tag/4.1.1
+- spaCy (3.2.1): https://github.com/explosion/spaCy/releases/tag/v3.2.1
-- spaCy medium sized models (3.0.0):
+- spaCy medium sized models (3.2.0):
-  - https://github.com/explosion/spacy-models/releases/tag/da_core_news_md-3.0.0
+  - https://github.com/explosion/spacy-models/releases/tag/de_core_news_md-3.2.0
-  - https://github.com/explosion/spacy-models/releases/tag/de_core_news_md-3.0.0
+  - https://github.com/explosion/spacy-models/releases/tag/en_core_web_md-3.2.0
-  - https://github.com/explosion/spacy-models/releases/tag/el_core_news_md-3.0.0
+  - https://github.com/explosion/spacy-models/releases/tag/it_core_news_md-3.2.0
-  - https://github.com/explosion/spacy-models/releases/tag/en_core_web_md-3.0.0
+  - https://github.com/explosion/spacy-models/releases/tag/nl_core_news_md-3.2.0
-  - https://github.com/explosion/spacy-models/releases/tag/es_core_news_md-3.0.0
+  - https://github.com/explosion/spacy-models/releases/tag/pl_core_news_md-3.2.0
-  - https://github.com/explosion/spacy-models/releases/tag/fr_core_news_md-3.0.0
+  - https://github.com/explosion/spacy-models/releases/tag/zh_core_web_md-3.2.0
-  - https://github.com/explosion/spacy-models/releases/tag/it_core_news_md-3.0.0
-  - https://github.com/explosion/spacy-models/releases/tag/nl_core_news_md-3.0.0
-  - https://github.com/explosion/spacy-models/releases/tag/pt_core_news_md-3.0.0
-  - https://github.com/explosion/spacy-models/releases/tag/ru_core_news_md-3.0.0
-  - https://github.com/explosion/spacy-models/releases/tag/zh_core_web_md-3.0.0
 
 
-## Use this image
+## Installation
 
-1. Create input and output directories for the pipeline.
+1. Install Docker and Python 3.
-``` bash
+2. Clone this repository: `git clone https://gitlab.ub.uni-bielefeld.de/sfb1288inf/nlp.git`
-mkdir -p /<my_data_location>/input /<my_data_location>/output
+3. Build the Docker image: `docker build -t gitlab.ub.uni-bielefeld.de:4567/sfb1288inf/nlp:v0.1.0 nlp`
-```
+4. Add the wrapper script (`wrapper/nlp` relative to this README file) to your `${PATH}`.
+5. Create working directories for the pipeline: `mkdir -p /<my_data_location>/{input,output}`.
 
-2. Place your text files inside `/<my_data_location>/input`. Files should all contain text of the same language.
 
-3. Start the pipeline process. Check the [Pipeline arguments](#pipeline-arguments) section for more details.
+## Use the Pipeline
-```
+
-# Option one: Use the wrapper script
+1. Place your plain text files inside `/<my_data_location>/input`. Files should all contain text of the same language.
-## Install the wrapper script (only on first run). Get it from https://gitlab.ub.uni-bielefeld.de/sfb1288inf/nlp/-/raw/1.0.0/wrapper/nlp, make it executeable and add it to your ${PATH}
+2. Clear your `/<my_data_location>/output` directory.
+3. Start the pipeline process. Check the pipeline help (`nlp --help`) for more details.
+```bash
 cd /<my_data_location>
-nlp -i input -l <language_code> -o output <optional_pipeline_arguments>
+nlp \
-
+  --input-dir input \
-# Option two: Classic Docker style
+  --output-dir output \
-docker run \
+  -m <model_code> <optional_pipeline_arguments>
-    --rm \
-    -it \
-    -u $(id -u $USER):$(id -g $USER) \
-    -v /<my_data_location>/input:/input \
-    -v /<my_data_location>/output:/output \
-    gitlab.ub.uni-bielefeld.de:4567/sfb1288inf/nlp:1.0.0 \
-        -i /input \
-        -l <language_code>
-        -o /output \
-        <optional_pipeline_arguments>
 ```
-
 4. Check your results in the `/<my_data_location>/output` directory.
-```
-
-### Pipeline arguments
-
-`--check-encoding`
-* If set, the pipeline tries to automatically determine the right encoding for
-your texts. Only use it if you are not sure that your input is provided in UTF-8.
-* default = False
-* required = False
-
-`-l languagecode`
-* Tells spaCy which language will be used.
-* options = da (Danish), de (German), el (Greek), en (English), es (Spanish), fr (French), it (Italian), nl (Dutch), pt (Portuguese), ru (Russian), zh (Chinese)
-* required = True
-
-`--nCores corenumber`
-* Sets the number of CPU cores being used during the NLP process.
-* default = min(4, multiprocessing.cpu_count())
-* required = False
-
-``` bash
-# Example with all arguments used
-docker run \
-    --rm \
-    -it \
-    -u $(id -u $USER):$(id -g $USER) \
-    -v "$HOME"/ocr/input:/input \
-    -v "$HOME"/ocr/output:/output \
-    gitlab.ub.uni-bielefeld.de:4567/sfb1288inf/nlp:1.0.0 \
-        -i /input \
-        -l en \
-        -o /output \
-        --check-encoding \
-        --nCores 8 \
-```

nlp

@@ -1,169 +1,286 @@
 #!/usr/bin/env python2.7
 # coding=utf-8
 
-"""A NLP pipeline for text file processing."""
+''' A NLP pipeline for text file processing. '''
-
+__version__ = '0.1.0'
-__author__ = 'Patrick Jentsch <p.jentsch@uni-bielefeld.de>,' \
-             'Stephan Porada <porada@posteo.de>'
-__version__ = '1.0.0'
 
 from argparse import ArgumentParser
 from pyflow import WorkflowRunner
-import multiprocessing
+import json
 import os
 import sys
 
 
-SPACY_MODELS = {'da': 'da_core_news_md',
+SPACY_MODELS = {
-                'de': 'de_core_news_md',
+    'de': 'de_core_news_md',
-                'el': 'el_core_news_md',
+    'en': 'en_core_web_md',
-                'en': 'en_core_web_md',
+    'it': 'it_core_news_md',
-                'es': 'es_core_news_md',
+    'nl': 'nl_core_news_md',
-                'fr': 'fr_core_news_md',
+    'pl': 'pl_core_news_md',
-                'it': 'it_core_news_md',
+    'zh': 'zh_core_web_md'
-                'nl': 'nl_core_news_md',
+}
-                'pt': 'pt_core_news_md',
-                'ru': 'ru_core_news_md',
-                'zh': 'zh_core_web_md'}
 
 
-def parse_args():
+class PipelineJob:
-    parser = ArgumentParser(description='NLP Pipeline utilizing spaCy.')
+    '''
-    parser.add_argument('-i', '--input-directory',
+    NLP pipeline job class.
-                        help='Input directory (only txt files get processed)',
-                        required=True)
-    parser.add_argument('-o', '--output-directory',
-                        help='Output directory',
-                        required=True)
-    parser.add_argument('-l', '--language',
-                        choices=SPACY_MODELS.keys(),
-                        required=True)
-    parser.add_argument('--check-encoding', action='store_true')
-    parser.add_argument('--log-dir')
-    parser.add_argument('--n-cores',
-                        default=min(4, multiprocessing.cpu_count()),
-                        help='total number of cores available', type=int)
-    parser.add_argument('--zip', help='Zips everything into one archive.')
-    return parser.parse_args()
 
-
+    Each plain text input file of the pipeline is represented as an NLP
-class NLPPipelineJob:
+    pipeline job, which holds all necessary information for the pipeline to
-    """An NLP pipeline job class
+    process it.
-
-    Each input file of the pipeline is represented as an NLP pipeline job,
-    which holds all necessary information for the pipeline to process it.
 
     Arguments:
     file -- Path to the file
-    output_dir -- Path to a directory, where job results a stored
+    output_dir -- Path to a directory, where job results are stored
-    intermediate_dir -- Path to a directory, where intermediate files are
+    '''
-                        stored.
-    """
 
     def __init__(self, file, output_dir):
         self.file = file
-        self.name = os.path.basename(file).rsplit('.', 1)[0]
+        self.name = os.path.basename(file)[:-4]
         self.output_dir = output_dir
 
 
-class NLPPipeline(WorkflowRunner):
+class NLPWorkflow(WorkflowRunner):
-    def __init__(self, input_dir, lang, output_dir, check_encoding, n_cores, zip):
+    def __init__(self, job, model, check_encoding=False, id_prefix=''):
+        self.job = job
+        self.model = model
+        self.check_encoding = check_encoding
+        self.id_prefix = id_prefix
+
+    def workflow(self):
+        '''
+        ' ##################################################
+        ' # spacy                                          #
+        ' ##################################################
+        '''
+        n_cores = 1
+        mem_mb = min(1024, self.getMemMb())
+        cmd = 'spacy-nlp'
+        cmd += ' --input-file "{}"'.format(self.job.file)
+        cmd += ' --output-file "{}"'.format(
+            os.path.join(self.job.output_dir, '{}.json'.format(self.job.name))
+        )
+        cmd += ' -m "{}"'.format(self.model)
+        if self.check_encoding:
+            cmd += ' --check-encoding'
+        cmd += ' --id-prefix "{}"'.format(self.id_prefix)
+        self.addTask(
+            'spacy',
+            command=cmd,
+            memMb=mem_mb,
+            nCores=n_cores
+        )
+
+
+class CreateVrtWorkflow(WorkflowRunner):
+    def __init__(self, job):
+        self.job = job
+
+    def workflow(self):
+        '''
+        ' ##################################################
+        ' # vrt-creator                                    #
+        ' ##################################################
+        '''
+        n_cores = 1
+        mem_mb = min(256, self.getMemMb())
+        cmd = 'vrt-creator'
+        cmd += ' --stand-off-data-file "{}"'.format(
+            os.path.join(self.job.output_dir, '{}.json'.format(self.job.name))
+        )
+        cmd += ' --text-file "{}"'.format(self.job.file)
+        cmd += ' --output-file "{}"'.format(
+            os.path.join(self.job.output_dir, '{}.vrt'.format(self.job.name))
+        )
+        self.addTask(
+            'vrt_creator',
+            command=cmd,
+            memMb=mem_mb,
+            nCores=n_cores
+        )
+
+
+class MainWorkflow(WorkflowRunner):
+    def __init__(
+        self,
+        input_dir,
+        model,
+        output_dir,
+        check_encoding=False,
+        id_prefix=''
+    ):
         self.input_dir = input_dir
-        self.lang = lang
+        self.model = model
         self.output_dir = output_dir
         self.check_encoding = check_encoding
-        self.n_cores = n_cores
+        self.id_prefix = id_prefix
-        self.output_dir = output_dir
+        self.jobs = []
-        if zip is None:
+
-            self.zip = zip
+    def collect_jobs(self):
-        else:
+        self.jobs = []
-            if zip.lower().endswith('.zip'):
+        for file in os.listdir(self.input_dir):
-                # Remove .zip file extension if provided
+            if os.path.isdir(os.path.join(self.input_dir, file)):
-                self.zip = zip[:-4]
+                continue
-                self.zip = self.zip if self.zip else 'output'
+            if not file.lower().endswith('.txt'):
-            else:
+                continue
-                self.zip = zip
+            job = PipelineJob(
-        self.jobs = collect_jobs(self.input_dir, self.output_dir)
+                os.path.join(self.input_dir, file),
+                os.path.join(self.output_dir, file)
+            )
+            self.jobs.append(job)
 
     def workflow(self):
         if not self.jobs:
             return
 
-        '''
+        # Create output and temporary directories
-        ' ##################################################
+        for job in self.jobs:
-        ' # setup output directory                         #
+            os.mkdir(job.output_dir)
-        ' ##################################################
-        '''
-        setup_output_directory_tasks = []
-        for i, job in enumerate(self.jobs):
-            cmd = 'mkdir'
-            cmd += ' -p'
-            cmd += ' "{}"'.format(job.output_dir)
-            lbl = 'setup_output_directory_-_{}'.format(i)
-            task = self.addTask(command=cmd, label=lbl)
-            setup_output_directory_tasks.append(task)
 
         '''
         ' ##################################################
-        ' # nlp                                 #
+        ' # nlp                                            #
         ' ##################################################
         '''
         nlp_tasks = []
-        n_cores = min(self.n_cores, max(1, int(self.n_cores / len(self.jobs))))
         for i, job in enumerate(self.jobs):
-            output_file = os.path.join(job.output_dir, '{}.vrt'.format(job.name))  # noqa
+            task = self.addWorkflowTask(
-            cmd = 'spacy-nlp'
+                'nlp_-_{}'.format(i),
-            cmd += ' -i "{}"'.format(job.file)
+                NLPWorkflow(
-            cmd += ' -l "{}"'.format(self.lang)
+                    job,
-            cmd += ' -o "{}"'.format(output_file)
+                    self.model,
-            if self.check_encoding:
+                    check_encoding=self.check_encoding,
-                cmd += ' --check-encoding'
+                    id_prefix=self.id_prefix
-            deps = 'setup_output_directory_-_{}'.format(i)
+                )
-            lbl = 'nlp_-_{}'.format(i)
+            )
-            task = self.addTask(command=cmd, dependencies=deps, label=lbl, nCores=n_cores)  # noqa
             nlp_tasks.append(task)
 
         '''
         ' ##################################################
-        ' # zip creation                                   #
+        ' # create vrt                                     #
         ' ##################################################
         '''
-        zip_creation_tasks = []
+        create_vrt_tasks = []
-        if self.zip is not None:
+        for i, job in enumerate(self.jobs):
-            cmd = 'cd "{}"'.format(self.output_dir)
+            task = self.addWorkflowTask(
-            cmd += ' && '
+                'create_vrt_-_{}'.format(i),
-            cmd += 'zip'
+                CreateVrtWorkflow(job),
-            cmd += ' -r'
+                dependencies='nlp_-_{}'.format(i)
-            cmd += ' "{}.zip" .'.format(self.zip)
+            )
-            cmd += ' -x "pyflow.data*"'
+            create_vrt_tasks.append(task)
-            cmd += ' -i "*.vrt"'
+
-            cmd += ' && '
+        self.waitForTasks()
-            cmd += 'cd -'
+        outputs = []
-            deps = nlp_tasks
+        for job in self.jobs:
-            lbl = 'zip_creation'
+            # Track output files
-            task = self.addTask(command=cmd, dependencies=deps, label=lbl)
+            relative_output_dir = os.path.relpath(
-            zip_creation_tasks.append(task)
+                job.output_dir,
+                start=self.output_dir
+            )
+            outputs.append(
+                {
+                    'description': 'JSON stand off data',
+                    'file': os.path.join(
+                        relative_output_dir,
+                        '{}.json'.format(job.name)
+                    ),
+                    'mimetype': 'application/json'
+                }
+            )
+            outputs.append(
+                {
+                    'description': 'CWB vrt file',
+                    'file': os.path.join(
+                        relative_output_dir,
+                        '{}.vrt'.format(job.name)
+                    ),
+                    'mimetype': 'application/vrt+xml'
+                }
+            )
+        with open(os.path.join(self.output_dir, 'outputs.json'), 'w') as f:
+            json.dump(outputs, f, indent=4)
 
 
-def collect_jobs(input_dir, output_dir):
+def parse_args():
-    jobs = []
+    parser = ArgumentParser(
-    for file in os.listdir(input_dir):
+        description='NLP pipeline for plain text file processing'
-        if os.path.isdir(os.path.join(input_dir, file)):
+    )
-            jobs += collect_jobs(os.path.join(input_dir, file),
+    parser.add_argument(
-                                 os.path.join(output_dir, file))
+        '-i', '--input-dir',
-        elif file.lower().endswith('.txt'):
+        help='Input directory',
-            jobs.append(NLPPipelineJob(os.path.join(input_dir, file),
+        required=True
-                                       os.path.join(output_dir, file)))
+    )
-    return jobs
+    parser.add_argument(
+        '-o', '--output-dir',
+        help='Output directory',
+        required=True
+    )
+    parser.add_argument(
+        '-m', '--model',
+        choices=SPACY_MODELS.keys(),
+        help='The model to be used',
+        required=True
+    )
+    parser.add_argument(
+        '--check-encoding',
+        action='store_true',
+        help='Check encoding of the input file, UTF-8 is used instead'
+    )
+    parser.add_argument(
+        '--id-prefix',
+        default='',
+        help='A prefix for all the ids within the stand off annotations'
+    )
+    parser.add_argument(
+        '--log-dir',
+        help='Logging directory (Default: --output-dir)'
+    )
+    parser.add_argument(
+        '--mem-mb',
+        help='Amount of system memory to be used '
+             '(Default: min(--n-cores * 1024, available system memory))',
+        type=int
+    )
+    parser.add_argument(
+        '--n-cores',
+        default=1,
+        help='Number of CPU threads to be used',
+        type=int
+    )
+    parser.add_argument(
+        '-v', '--version',
+        action='version',
+        help='Returns the current version of the NLP pipeline',
+        version='%(prog)s {}'.format(__version__)
+    )
+    args = parser.parse_args()
+
+    # Set some tricky default values and check for insufficient input
+    if args.log_dir is None:
+        args.log_dir = args.output_dir
+    if args.n_cores < 1:
+        raise Exception('--n-cores must be greater or equal 1')
+    if args.mem_mb is None:
+        max_mem_mb = int(os.popen('free -t -m').readlines()[-1].split()[1:][0])
+        args.mem_mb = min(args.n_cores * 1024, max_mem_mb)
+    if args.mem_mb < 1024:
+        raise Exception('--mem-mb must be greater or equal 1024')
+    return args
 
 
 def main():
     args = parse_args()
-    nlp_pipeline = NLPPipeline(args.input_directory, args.language,
+    main_workflow = MainWorkflow(
-                               args.output_directory, args.check_encoding,
+        args.input_dir,
-                               args.n_cores, args.zip)
+        args.model,
-    retval = nlp_pipeline.run(
+        args.output_dir,
-        dataDirRoot=(args.log_dir or args.output_directory),
+        check_encoding=args.check_encoding,
+        id_prefix=args.id_prefix
+    )
+    main_workflow.collect_jobs()
+    retval = main_workflow.run(
+        dataDirRoot=args.log_dir,
+        memMb=args.mem_mb,
         nCores=args.n_cores
     )
     sys.exit(retval)

packages/stand-off-data-py/setup.py

@@ -0,0 +1,14 @@
+import setuptools
+
+setuptools.setup(
+    name='Stand off data',
+    author='Patrick Jentsch',
+    author_email='p.jentsch@uni-bielefeld.de',
+    description='A python library to handle stand off data.',
+    py_modules=['stand_off_data'],
+    classifiers=[
+        'Programming Language :: Python :: 3',
+        'Operating System :: OS Independent',
+    ],
+    python_requires='>=3.5'
+)

packages/stand-off-data-py/stand_off_data.py

@@ -0,0 +1,282 @@
+from xml.sax.saxutils import escape
+
+
+class StandOffData:
+    def __init__(self, attrs={}):
+        self.meta = attrs.get('meta', {})
+        self.lookup = {}
+        for x in attrs.get('tags', []):
+            self.add_tag_definition(x)
+        self.annotations = [
+            TagAnnotation(x, self.lookup)
+            for x in attrs.get('annotations', [])
+        ]
+
+    def add_tag_definition(self, attrs):
+        tag_definition = TagDefinition(attrs)
+        if tag_definition.id in self.lookup:
+            raise Exception(f'Tag id already in use: {self.to_dict()}')
+        self.lookup[tag_definition.id] = tag_definition
+
+    def to_dict(self):
+        return {
+            'meta': self.meta,
+            'lookup': {k: v.to_dict() for k, v in self.lookup.items()},
+            'annotations': [x.to_dict() for x in self.annotations]
+        }
+
+    def to_vrt(self, text):
+        # Devide annotations into CWB's verticalized text format (.vrt) logic
+        p_attrs = []    # positional attributes
+        s_attrs = []    # structural attributes
+        for annotation in self.annotations:
+            if annotation.name == 'token':
+                p_attrs.append(annotation)
+            else:
+                s_attrs.append(annotation)
+        # Sort annotations, necessary for the next checks
+        p_attrs.sort()
+        s_attrs.sort()
+        # Check for p_attr<->p_attr overlap
+        for i, p_attr in enumerate(p_attrs[:-1]):
+            next_p_attr = p_attrs[i + 1]
+            # Check if first_p_attr starts/ends within second_p_attr
+            if ((p_attr.start >= next_p_attr.start) and (p_attr.start < next_p_attr.end)  # noqa
+                or (p_attr.end > next_p_attr.start) and (p_attr.end <= next_p_attr.end)):  # noqa
+                raise Exception(
+                    'Positional attribute overlaps another: '
+                    f'{p_attr.to_dict()}<->{next_p_attr.to_dict()}'
+                )
+        # Check for s_attr<->p_attr overlap
+        for i, s_attr in enumerate(s_attrs):
+            for p_attr in p_attrs:
+                # Check if s_attr starts within p_attr
+                if s_attr.start > p_attr.start and s_attr.start < p_attr.end:
+                    # Change s_attr start to p_attr's start
+                    s_attrs[i].start = p_attr.start
+                # Check if s_attr ends within p_attr
+                if s_attr.end < p_attr.end and s_attr.end > p_attr.start:
+                    # Change s_attr end to p_attr's end
+                    s_attrs[i].end = p_attr.end
+                # Check if s_attr starts/ends before/after p_attr
+                if p_attr.start >= s_attr.end or p_attr.end <= s_attr.start:
+                    # No further Checking needed (because p_attrs are sorted)
+                    break
+        p_attr_buffer = {}
+        for i, p_attr in enumerate(p_attrs):
+            p_attr_buffer[p_attr.start] = i
+        s_attr_start_buffer = {}
+        s_attr_end_buffer = {}
+        for i, s_attr in enumerate(s_attrs):
+            if s_attr.start in s_attr_start_buffer:
+                s_attr_start_buffer[s_attr.start].append(i)
+            else:
+                s_attr_start_buffer[s_attr.start] = [i]
+            if s_attr.end in s_attr_end_buffer:
+                s_attr_end_buffer[s_attr.end].insert(0, i)
+            else:
+                s_attr_end_buffer[s_attr.end] = [i]
+        vrt = ''
+        vrt += '<text>\n'
+        current_position = 0
+        text_len = len(text)
+        # As long as we have something in our buffers we process it
+        while current_position <= text_len:
+            # s_attr endings
+            # for k in {k: v for k, v in s_attr_end_buffer.items() if k <= current_position}:  # noqa
+            if current_position in s_attr_end_buffer:
+                # s_attr_indexes = s_attr_end_buffer.pop(k)
+                s_attr_indexes = s_attr_end_buffer.pop(current_position)
+                for s_attr_index in s_attr_indexes:
+                    s_attr = s_attrs[s_attr_index]
+                    vrt += f'</{escape(s_attr.name)}>\n'
+            # s_attrs starts
+            # for k in {k: v for k, v in s_attr_start_buffer.items() if k <= current_position}:  # noqa
+            if current_position in s_attr_start_buffer:
+                # s_attr_indexes = s_attr_start_buffer.pop(k)
+                s_attr_indexes = s_attr_start_buffer.pop(current_position)
+                for s_attr_index in s_attr_indexes:
+                    s_attr = s_attrs[s_attr_index]
+                    vrt += f'<{escape(s_attr.name)}'
+                    for property in s_attr.properties:
+                        vrt += f' {escape(property.name)}="{escape(str(property.value))}"'  # noqa
+                    vrt += '>\n'
+            # p_attrs
+            if current_position not in p_attr_buffer:
+                current_position += 1
+                continue
+            p_attr_index = p_attr_buffer.pop(current_position)
+            p_attr = p_attrs[p_attr_index]
+            if text[p_attr.start:p_attr.end].isspace():
+                current_position = p_attr.end
+                continue
+            _p_attr = {
+                'lemma': 'None',
+                'pos': 'None',
+                'simple_pos': 'None',
+                'word': 'None'
+            }
+            for property in p_attr.properties:
+                if property.name not in _p_attr:
+                    continue
+                _p_attr[property.name] = escape(str(property.value))
+            _p_attr['word'] = escape(text[p_attr.start:p_attr.end])
+            vrt += '{word}\t{pos}\t{lemma}\t{simple_pos}\n'.format(**_p_attr)
+            current_position = p_attr.end
+        vrt += '</text>\n'
+        return vrt
+
+
+class TagAnnotation:
+    def __init__(self, attrs, lookup):
+        self.lookup = lookup
+        self.tag_id = attrs['tag_id']
+        self.start = attrs['start']
+        self.end = attrs['end']
+        self.properties = [
+            PropertyAnnotation(x, self.lookup[self.tag_id].properties)
+            for x in attrs.get('properties', [])
+        ]
+        ''' Sanity checks '''
+        if self.tag_id not in self.lookup:
+            raise Exception(f'Unknown tag: {self.to_dict()}')
+        if self.end < self.start:
+            raise Exception(f'Annotation end less then start: {self.to_dict()}')  # noqa
+        # property_ids = [x.property_id for x in self.properties]
+        # for required_property_id, required_property in self.lookup[self.tag_id].required_properties.items():  # noqa
+        #     if required_property_id not in property_ids:
+        #         raise Exception(
+        #             f'Missing required property: {required_property.to_dict()}'
+        #         )
+
+    @property
+    def name(self):
+        return self.lookup[self.tag_id].name
+
+    def to_dict(self):
+        return {
+            'tag_id': self.tag_id,
+            'start': self.start,
+            'end': self.end,
+            'properties': [x.to_dict() for x in self.properties]
+        }
+
+    def __lt__(self, other):
+        if self.start == other.start:
+            if self.name == 'token' and other.name != 'token':
+                return False
+            elif self.name != 'token' and other.name == 'token':
+                return True
+            else:
+                return self.end > other.end
+        else:
+            return self.start < other.start
+
+    def __le__(self, other):
+        if self.start == other.start:
+            if self.name == 'token' and other.name != 'token':
+                return False
+            elif self.name != 'token' and other.name == 'token':
+                return True
+            else:
+                return self.end >= other.end
+        else:
+            return self.start <= other.start
+
+    def __eq__(self, other):
+        if self.start == other.start:
+            if self.name == 'token' and other.name != 'token':
+                return False
+            elif self.name != 'token' and other.name == 'token':
+                return False
+            else:
+                return self.end == other.end
+        else:
+            return False
+
+    def __ne__(self, other):
+        return not self == other
+
+    def __gt__(self, other):
+        return not self <= other
+
+    def __ge__(self, other):
+        return not self < other
+
+
+class PropertyAnnotation:
+    def __init__(self, attrs, lookup):
+        self.lookup = lookup
+        self.property_id = attrs['property_id']
+        self.value = attrs['value']
+        # TODO: Process attrs['possibleValues'] as self.labels (no id?)
+        ''' Sanity checks '''
+        if self.property_id not in self.lookup:
+            raise Exception(f'Unknown property: {self.to_dict()}')
+
+    @property
+    def name(self):
+        return self.lookup[self.property_id].name
+
+    def to_dict(self):
+        return {
+            'property_id': self.property_id,
+            'tag_id': self.tag_id,
+            'value': self.value
+        }
+
+
+class TagDefinition:
+    def __init__(self, attrs):
+        self.id = attrs['id']
+        self.name = attrs['name']
+        self.description = attrs.get('description', '')
+        self.properties = {}
+        for x in attrs.get('properties', []):
+            self.add_property_definition(x)
+
+    def add_property_definition(self, attrs):
+        property_definition = PropertyDefinition(attrs)
+        if property_definition.id in self.properties:
+            raise Exception(
+                f'Property id already in use: {property_definition.to_dict()}')
+        self.properties[property_definition.id] = property_definition
+
+    # @property
+    # def required_properties(self):
+    #     return {property.id: property for property in self.properties.values()
+    #             if property.is_required}
+
+    def to_dict(self):
+        return {
+            'id': self.id,
+            'name': self.name,
+            'description': self.description,
+            'properties': {k: v.to_dict() for k, v in self.properties.items()}
+        }
+
+
+class PropertyDefinition:
+    def __init__(self, attrs):
+        self.id = attrs['id']
+        self.name = attrs['name']
+        self.description = attrs.get('description', '')
+        self.flags = attrs.get('flags', [])
+        self.labels = attrs.get('labels', [])
+
+    # @property
+    # def is_required(self):
+    #     return 'required' in self.flags
+
+    @property
+    def has_multiple_values(self):
+        return 'multiple' in self.flags
+
+    def to_dict(self):
+        return {
+            'id': self.id,
+            'name': self.name,
+            'description': self.description,
+            'flags': self.flags,
+            'labels': self.labels
+        }

packages/stand-off-data-py/stand_off_data/__init__.py

@@ -0,0 +1,2 @@
+# flake8: noqa
+from .models import StandOffData

spacy-nlp

@@ -2,118 +2,352 @@
 # coding=utf-8
 
 from argparse import ArgumentParser
-from xml.sax.saxutils import escape
 import chardet
 import hashlib
+import json
 import os
 import spacy
 import textwrap
+import uuid
 
 
-SPACY_MODELS = {'da': 'da_core_news_md',
+spacy_models = {
-                'de': 'de_core_news_md',
+    spacy.info(pipeline)['lang']: pipeline
-                'el': 'el_core_news_md',
+    for pipeline in spacy.info()['pipelines']
-                'en': 'en_core_web_md',
+}
-                'es': 'es_core_news_md',
-                'fr': 'fr_core_news_md',
-                'it': 'it_core_news_md',
-                'nl': 'nl_core_news_md',
-                'pt': 'pt_core_news_md',
-                'ru': 'ru_core_news_md',
-                'zh': 'zh_core_web_md'}
 
 
-SPACY_MODELS_VERSION = os.environ.get('SPACY_MODELS_VERSION')
-SPACY_VERSION = os.environ.get('SPACY_VERSION')
-
 # Parse the given arguments
-parser = ArgumentParser(description=('Tag a text file with spaCy and save it '
+parser = ArgumentParser(
-                                     'as a verticalized text file.'))
+    description='Create annotations for a given plain txt file'
-parser.add_argument('-i', '--input', metavar='txt-sourcefile', required=True)
+)
-parser.add_argument('-o', '--output', metavar='vrt-destfile', required=True)
+parser.add_argument(
-parser.add_argument('-l', '--language', choices=SPACY_MODELS.keys(), required=True)  # noqa
+    '-i', '--input-file',
-parser.add_argument('--check-encoding', action='store_true')
+    help='Input file',
+    required=True
+)
+parser.add_argument(
+    '-o', '--output-file',
+    help='Output file',
+    required=True
+)
+parser.add_argument(
+    '-m', '--model',
+    choices=spacy_models.keys(),
+    help='The model to be used',
+    required=True
+)
+parser.add_argument(
+    '-c', '--check-encoding',
+    action='store_true',
+    help='Check encoding of the input file, UTF-8 is used instead'
+)
+parser.add_argument(
+    '--id-prefix',
+    default='',
+    help='A prefix for all the ids within the stand off annotations'
+)
 args = parser.parse_args()
 
 
-# If requested: Check the encoding of the text contents from the input file
+def generate_id(name):
-# Else: Use utf-8
+    return f'{args.id_prefix}{uuid.uuid3(uuid.NAMESPACE_DNS, name)}'
-if args.check_encoding:
-    with open(args.input, "rb") as input_file:
-        bytes = input_file.read()
-        encoding = chardet.detect(bytes)['encoding']
-else:
-    encoding = 'utf-8'
 
 
-# hashing in chunks to avoid full RAM with huge files.
+with open(args.input_file, "rb") as input_file:
-with open(args.input, 'rb') as input_file:
+    if args.check_encoding:
-    source_md5 = hashlib.md5()
+        encoding = chardet.detect(input_file.read())['encoding']
-    for chunk in iter(lambda: input_file.read(128 * source_md5.block_size), b''):
+    else:
-        source_md5.update(chunk)
+        encoding = 'utf-8'
-    source_md5 = source_md5.hexdigest()
+    input_file.seek(0)
+    text_md5 = hashlib.md5()
+    for chunk in iter(lambda: input_file.read(128 * text_md5.block_size), b''):
+        text_md5.update(chunk)
 
 # Load the text contents from the input file
-with open(args.input, encoding=encoding) as input_file:
+with open(args.input_file, encoding=encoding) as input_file:
-    text = input_file.read()
+    # spaCy NLP is limited to strings with a maximum of 1 million characters at
-    # spaCys NLP is limited to strings with maximum 1 million characters at
     # once. So we split it into suitable chunks.
-    text_chunks = textwrap.wrap(text, 1000000, break_long_words=False)
+    text_chunks = textwrap.wrap(
-    # the text variable potentially occupies a lot of system memory and is no
+        input_file.read(),
-    # longer needed...
+        1000000,
-    del text
+        break_long_words=False,
+        break_on_hyphens=False,
+        drop_whitespace=False,
+        expand_tabs=False,
+        replace_whitespace=False
+    )
 
+model_name = spacy_models[args.model]
+nlp = spacy.load(model_name)
 
-# Setup the spaCy toolkit by loading the chosen language model
+meta = {
-model = SPACY_MODELS[args.language]
+    'generator': {
-nlp = spacy.load(model)
+        'name': 'nopaque spacy NLP',
+        'version': '0.1.0',
+        'arguments': {
+            'check_encoding': args.check_encoding,
+            'model': args.model
+        }
+    },
+    'file': {
+        'encoding': encoding,
+        'md5': text_md5.hexdigest(),
+        'name': os.path.basename(args.input_file)
+    }
+}
 
+tags = []
+token = {
+    'id': generate_id('token'),
+    'name': 'token',
+    'description': 'An individual token — i.e. a word, punctuation symbol, whitespace, etc.',  # noqa
+    'properties': []
+}
+# TODO: Check if all languages support token.sentiment
+token['properties'].append(
+    {
+        'id': generate_id('token.sentiment'),
+        'name': 'sentiment',
+        'description': 'A scalar value indicating the positivity or negativity of the token.'  # noqa
+    }
+)
+if nlp.has_pipe('lemmatizer'):
+    token['properties'].append(
+        {
+            'id': generate_id('token.lemma'),
+            'name': 'lemma',
+            'description': 'The base form of the word'
+        }
+    )
+if nlp.has_pipe('morphologizer') or nlp.has_pipe('tagger'):
+    token['properties'].append(
+        {
+            'id': generate_id('token.simple_pos'),
+            'name': 'simple_pos',
+            'description': 'The simple UPOS part-of-speech tag',
+            'labels': [
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'ADJ',
+                    'description': 'adjective'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'ADP',
+                    'description': 'adposition'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'ADV',
+                    'description': 'adverb'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'AUX',
+                    'description': 'auxiliary verb'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'CONJ',
+                    'description': 'coordinating conjunction'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'DET',
+                    'description': 'determiner'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'INTJ',
+                    'description': 'interjection'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'NOUN',
+                    'description': 'noun'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'NUM',
+                    'description': 'numeral'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'PART',
+                    'description': 'particle'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'PRON',
+                    'description': 'pronoun'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'PROPN',
+                    'description': 'proper noun'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'PUNCT',
+                    'description': 'punctuation'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'SCONJ',
+                    'description': 'subordinating conjunction'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'SYM',
+                    'description': 'symbol'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'VERB',
+                    'description': 'verb'
+                },
+                {
+                    'id': generate_id('token.simple_pos=ADJ'),
+                    'name': 'X',
+                    'description': 'other'
+                }
+            ]
+        }
+    )
+if nlp.has_pipe('tagger'):
+    token['properties'].append(
+        {
+            'id': generate_id('token.pos'),
+            'name': 'pos',
+            'description': 'The detailed part-of-speech tag',
+            'labels': [
+                {
+                    'id': generate_id(f'token.pos={label}'),
+                    'name': label,
+                    'description': spacy.explain(label) or ''
+                } for label in spacy.info(model_name)['labels']['tagger']
+            ]
+        }
+    )
+if nlp.has_pipe('ner') or nlp.has_pipe('entity_ruler'):
+    tags.append(
+        {
+            'id': generate_id('ent'),
+            'name': 'ent',
+            'description': 'Encodes the start and end of a named entity',
+            'properties': [
+                {
+                    'id': generate_id('ent.type'),
+                    'name': 'type',
+                    'description': 'Label indicating the type of the entity',
+                    'labels': [
+                        {
+                            'id': generate_id('ent.type={}'.format(label)),
+                            'name': label,
+                            'description': spacy.explain(label) or ''
+                        } for label in spacy.info(model_name)['labels']['ner']
+                    ]
+                }
+            ]
+        }
+    )
+if nlp.has_pipe('parser') or nlp.has_pipe('senter') or nlp.has_pipe('sentencizer'):  # noqa
+    # TODO: Check if all languages support sent.sentiment
+    tags.append(
+        {
+            'id': generate_id('s'),
+            'name': 's',
+            'description': 'Encodes the start and end of a sentence',
+            'properties': [
+                {
+                    'id': generate_id('s.sentiment'),
+                    'name': 'sentiment',
+                    'description': 'A scalar value indicating the positivity or negativity of the sentence.'  # noqa
+                }
+            ]
+        }
+    )
+tags.append(token)
 
-# Create the output file in verticalized text format
+annotations = []
-# See: http://cwb.sourceforge.net/files/CWB_Encoding_Tutorial/node3.html
-output_file_original_filename = args.output
-output_file_stand_off_filename = args.output.replace('.vrt', '.stand-off.vrt')
-common_xml = ('<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\n'
-              + '<corpus>\n'
-              + '<text>\n'
-              + '<nlp name="spaCy:{}"\n'.format(SPACY_VERSION)
-              + '     model="{}:{}"\n'.format(model, SPACY_MODELS_VERSION)
-              + '     source-md5="{}" />\n'.format(source_md5))
 
-with open(output_file_original_filename, 'w+') as output_file_original, \
+chunk_offset = 0
-     open(output_file_stand_off_filename, 'w+') as output_file_stand_off:
+while text_chunks:
-
+    text_chunk = text_chunks.pop(0)
-    output_file_original.write(common_xml)
+    doc = nlp(text_chunk)
-    output_file_stand_off.write(common_xml)
+    if hasattr(doc, 'ents'):
-    text_offset = 0
+        for ent in doc.ents:
-    for text_chunk in text_chunks:
+            annotation = {
-        doc = nlp(text_chunk)
+                'start': ent.start_char + chunk_offset,
+                'end': ent.end_char + chunk_offset,
+                'tag_id': generate_id('ent'),
+                'properties': [
+                    {
+                        'property_id': generate_id('ent.type'),
+                        'value': ent.label_
+                    }
+                ]
+            }
+            annotations.append(annotation)
+    if hasattr(doc, 'sents'):
         for sent in doc.sents:
-            output_file_original.write('<s>\n')
+            annotation = {
-            output_file_stand_off.write('<s>\n')
+                'start': sent.start_char + chunk_offset,
-            space_flag = False
+                'end': sent.end_char + chunk_offset,
-            # Skip whitespace tokens
+                'tag_id': generate_id('s'),
-            sent_no_space = [token for token in sent
+                'properties': []
-                             if not token.text.isspace()]
+            }
-            # No space variant for cwb original .vrt file input.
+            if hasattr(sent, 'sentiment'):
-            for token in sent_no_space:
+                annotation['properties'].append(
-                output_file_original.write('{}'.format(escape(token.text))
+                    {
-                                           + '\t{}'.format(escape(token.lemma_))
+                        'property_id': generate_id('s.sentiment'),
-                                           + '\t{}'.format(token.pos_)
+                        'value': sent.sentiment
-                                           + '\t{}'.format(token.tag_)
+                    }
-                                           + '\t{}\n'.format(token.ent_type_ or 'NULL'))
+                )
-            # Stand off variant with spaces.
+            annotations.append(annotation)
-            for token in sent:
+    for token in doc:
-                token_start = token.idx + text_offset
+        annotation = {
-                token_end = token.idx + len(token.text) + text_offset
+            'start': token.idx + chunk_offset,
-                output_file_stand_off.write('{}:{}'.format(token_start,
+            'end': token.idx + len(token.text) + chunk_offset,
-                                                           token_end)
+            'tag_id': generate_id('token'),
-                                            + '\t{}'.format(escape(token.lemma_))
+            'properties': []
-                                            + '\t{}'.format(token.pos_)
+        }
-                                            + '\t{}'.format(token.tag_)
+        if hasattr(token, 'lemma_'):
-                                            + '\t{}\n'.format(token.ent_type_ or 'NULL'))
+            annotation['properties'].append(
-            output_file_original.write('</s>\n')
+                {
-            output_file_stand_off.write('</s>\n')
+                    'property_id': generate_id('token.lemma'),
-        text_offset = token_end + 1
+                    'value': token.lemma_
-    output_file_original.write('</text>\n</corpus>')
+                }
-    output_file_stand_off.write('</text>\n</corpus>')
+            )
+        if hasattr(token, 'pos_'):
+            annotation['properties'].append(
+                {
+                    'property_id': generate_id('token.simple_pos'),
+                    'value': token.pos_
+                }
+            )
+        if hasattr(token, 'sentiment'):
+            annotation['properties'].append(
+                {
+                    'property_id': generate_id('token.sentiment'),
+                    'value': token.sentiment
+                }
+            )
+        if hasattr(token, 'tag_'):
+            annotation['properties'].append(
+                {
+                   'property_id': generate_id('token.pos'),
+                   'value': token.tag_
+                }
+            )
+        annotations.append(annotation)
+    chunk_offset += len(text_chunk)
+    text_chunk = None
+
+with open(args.output_file, 'w') as output_file:
+    json.dump(
+        {'meta': meta, 'tags': tags, 'annotations': annotations},
+        output_file,
+        indent=4
+    )

vrt-creator

@@ -0,0 +1,45 @@
+#!/usr/bin/env python3.7
+# coding=utf-8
+
+from argparse import ArgumentParser
+from stand_off_data import StandOffData
+import hashlib
+import json
+
+
+parser = ArgumentParser(
+    description='Convert plain text and JSON stand off to a CWB vrt file'
+)
+parser.add_argument(
+    '-s', '--stand-off-data-file',
+    help='JSON stand off data input file',
+    required=True
+)
+parser.add_argument(
+    '-t', '--text-file',
+    help='Plain text input file',
+    required=True
+)
+parser.add_argument(
+    '-o', '--output-file',
+    help='Output file',
+    required=True
+)
+args = parser.parse_args()
+
+
+with open(args.stand_off_data_file) as stand_of_data_file:
+    stand_off_data = StandOffData(json.load(stand_of_data_file))
+
+with open(args.text_file, "rb") as text_file:
+    text_md5 = hashlib.md5()
+    for chunk in iter(lambda: text_file.read(128 * text_md5.block_size), b''):
+        text_md5.update(chunk)
+    if text_md5.hexdigest() != stand_off_data.meta['file']['md5']:
+        raise Exception('md5 not equal')
+
+with open(args.text_file, encoding=stand_off_data.meta['file']['encoding']) as text_file:  # noqa
+    text = text_file.read()
+
+with open(args.output_file, 'w') as vrt_file:
+    vrt_file.write(stand_off_data.to_vrt(text))

wrapper/nlp

@@ -4,30 +4,35 @@
 from argparse import ArgumentParser
 import os
 import subprocess
+import sys
 
-CONTAINER_IMAGE = 'gitlab.ub.uni-bielefeld.de:4567/sfb1288inf/nlp:1.0.0'
+CONTAINER_IMAGE = 'gitlab.ub.uni-bielefeld.de:4567/sfb1288inf/nlp:v0.1.0'
 CONTAINER_INPUT_DIR = '/input'
 CONTAINER_OUTPUT_DIR = '/output'
+CONTAINER_LOG_DIR = '/logs'
 UID = str(os.getuid())
 GID = str(os.getgid())
 
 parser = ArgumentParser(add_help=False)
-parser.add_argument('-i', '--input-directory')
+parser.add_argument('-i', '--input-dir')
-parser.add_argument('-o', '--output-directory')
+parser.add_argument('-o', '--output-dir')
+parser.add_argument('--log-dir')
 args, remaining_args = parser.parse_known_args()
 
-cmd = ['docker', 'run', '--rm', '-it', '-u', '{}:{}'.format(UID, GID)]
+cmd = ['docker', 'run', '--rm', '-it', '-u', f'{UID}:{GID}']
-if args.output_directory is not None:
+if args.input_dir is not None:
-    cmd += ['-v', '{}:{}'.format(os.path.abspath(args.output_directory),
+    mapping = f'{os.path.abspath(args.input_dir)}:{CONTAINER_INPUT_DIR}'
-                                 CONTAINER_OUTPUT_DIR)]
+    cmd += ['-v', mapping]
-    remaining_args.insert(0, CONTAINER_OUTPUT_DIR)
+    remaining_args += ['-i', CONTAINER_INPUT_DIR]
-    remaining_args.insert(0, '-o')
+if args.output_dir is not None:
-if args.input_directory is not None:
+    mapping = f'{os.path.abspath(args.output_dir)}:{CONTAINER_OUTPUT_DIR}'
-    cmd += ['-v', '{}:{}'.format(os.path.abspath(args.input_directory),
+    cmd += ['-v', mapping]
-                                 CONTAINER_INPUT_DIR)]
+    remaining_args += ['-o', CONTAINER_OUTPUT_DIR]
-    remaining_args.insert(0, CONTAINER_INPUT_DIR)
+if args.log_dir is not None:
-    remaining_args.insert(0, '-i')
+    mapping = '{os.path.abspath(args.log_dir)}:{CONTAINER_LOG_DIR}'
+    cmd += ['-v', mapping]
+    remaining_args += ['--log-dir', CONTAINER_LOG_DIR]
 cmd.append(CONTAINER_IMAGE)
 cmd += remaining_args
 
-subprocess.run(cmd)
+sys.exit(subprocess.run(cmd).returncode)