[INFRA] set up github action to detect latin phrases

.github/workflows/no-bad-latin.yml

@@ -0,0 +1,47 @@
+# Give your test a name!
+name: Check for Latin Phrases
+
+# Decide when to run the tests
+#
+# This configuration sets the test to run on pushes to master
+# and on pull requests that are opened to master
+on:
+  push:
+    branches:
+      - master
+  pull_request:
+    branches:
+      - master
+
+# Set up the Continuous Integration job
+jobs:
+  latin-phrases:
+    # Run on the latest Ubuntu distribution
+    runs-on: ubuntu-latest
+    # This section collects together the steps involved in running the test
+    steps:
+    # Checkout the repository. Relies on another GH-Action.
+    - uses: actions/checkout@v2
+    # Set up the Python version. Relies on another GH-Action.
+    - name: Setup Python 3.7
+      uses: actions/setup-python@v1
+      with:
+        python-version: 3.7
+    # Install Python dependencies
+    - name: Install dependencies
+      working-directory: ./tools
+      run: |
+        python -m pip install --upgrade pip
+        python -m pip install -r requirements.txt
+    # Run a Python script
+    - name: Run Python script to check for latin phrases - Master
+      working-directory: ./tools
+      if: github.event_name == 'push' && github.ref == 'refs/heads/master'
+      run: |
+        python no-bad-latin.py
+
+    - name: Run Python script to check for latin phrases - Pull Request
+      working-directory: ./tools
+      if: github.event.pull_request
+      run: |
+        python no-bad-latin.py --pull-request ${{ github.event.pull_request.number }}

tools/no-bad-latin.py

@@ -0,0 +1,156 @@
+import os
+import re
+import argparse
+from pull_files import filter_files
+
+HERE = os.getcwd()
+ABSOLUTE_HERE = os.path.dirname(HERE)
+IGNORE_LIST = [ "_config.yml", "style.md", "contributors-record.md"]
+
+
+def parse_args():
+    """Construct command line interface for parsing Pull Request number"""
+    DESCRIPTION = "Script to check for latin phrases in Markdown files"
+    parser = argparse.ArgumentParser(description=DESCRIPTION)
+
+    parser.add_argument(
+        "--pull-request",
+        type=str,
+        default=None,
+        help="If the script is being run on a Pull Request, parse the PR number",
+    )
+
+    return parser.parse_args()
+
+
+def remove_comments(text_string):
+    """Function to omit  html comment identifiers in a text string using
+	regular expression matches
+
+	Arguments:
+		text_string {string} -- The text to be matched
+
+	Returns:
+		{string} -- The input text string with html comments removed
+	"""
+    p = re.sub("(?s)<!--(.*?)-->", "", text_string)
+    return p
+
+
+def get_lines(text_string, sub_string):
+    """Get individual lines in a text file
+
+	Arguments:
+		text_string {string} -- The text string to test
+		sub_string {string} -- The conditional string to perform splitting on
+
+	Returns:
+		{list} -- A list of split strings
+	"""
+    lines = [line for line in text_string.split("\n") if sub_string in line]
+    return lines
+
+
+def construct_error_message(files_dict):
+    """Function to construct an error message pointing out where bad latin
+	phrases appear in lines of text
+
+	Arguments:
+		files_dict {dictionary} -- Dictionary of failing files containing the
+		                           bad latin phrases and offending lines
+
+	Returns:
+		{string} -- The error message to be raised
+	"""
+    error_message = ["Bad latin found in the following files:\n"]
+
+    for file in files_dict.keys():
+        error_message.append(
+            f"{file}:\t{files_dict[file]['latin_type']}\tfound in line\t[{files_dict[file]['line']}]\n"
+        )
+
+    return "\n".join(error_message)
+
+
+def read_and_check_files(files):
+    """Function to read in files, remove html comments and check for bad latin
+	phrases
+
+	Arguments:
+		files {list} -- List of filenames to be checked
+
+	Returns:
+		{dict} -- Dictionary: Top level keys are absolute filepaths to files
+		          that failed the check. Each of these has two keys:
+				  'latin_type' containing the unwanted latin phrase, and 'line'
+				  containing the offending line.
+	"""
+    failing_files = {}
+    bad_latin = [
+        "i.e.", "i.e ", " ie ", 
+        "e.g.", "e.g ",  
+        "e.t.c.", " etc", "et cetera"]
+
+    for filename in files:
+        if os.path.basename(filename) in IGNORE_LIST:
+            pass
+        else:
+            try:
+                with open(
+                os.path.join(ABSOLUTE_HERE, filename), encoding="utf8",
+                errors="ignore") as f:
+                    text = f.read()
+                    text = remove_comments(text)
+
+                    for latin_type in bad_latin:
+                        if latin_type in text.lower():
+                            lines = get_lines(text.lower(), latin_type)
+                            for line in lines:
+                                failing_files[os.path.abspath(filename)] = {
+                                    "latin_type": latin_type,
+                                    "line": line,
+                                }
+            except FileNotFoundError:
+                pass
+
+    return failing_files
+
+
+def get_all_files(directory=os.path.join(ABSOLUTE_HERE, "src")):
+    """Get a list of files to be checked. Ignores image, javacript, css files.
+
+	Keyword Arguments:
+		directory {string} -- The directory containing the files to check
+
+	Returns:
+		{list} -- List of files to check
+	"""
+    files = []
+    filetypes_to_ignore = (".png", ".jpg", ".js", ".css")
+
+    for rootdir, _, filenames in os.walk(directory):
+        for filename in filenames:
+            if not filename.endswith(filetypes_to_ignore):
+                files.append(os.path.join(rootdir, filename))
+
+    return files
+
+
+def main():
+    """Main function"""
+    args = parse_args()
+
+    if args.pull_request is not None:
+        files = filter_files(args.pull_request)
+    else:
+        files = get_all_files()
+
+    failing_files = read_and_check_files(files)
+
+    if bool(failing_files):
+        error_message = construct_error_message(failing_files)
+        raise Exception(error_message)
+
+
+if __name__ == "__main__":
+    main()

tools/pull_files.py

@@ -0,0 +1,69 @@
+"""
+Script to pull changed files in a Pull Request using a GET resquest to the
+GitHub API.
+"""
+import requests
+import argparse
+
+
+def parse_args():
+    """Construct the command line interface for the script"""
+    DESCRIPTION = "Script to pull changed files in a Pull Request using a GET resquest to the GitHub API."
+    parser = argparse.ArgumentParser(description=DESCRIPTION)
+
+    parser.add_argument(
+        "--pull-request",
+        type=str,
+        default=None,
+        help="If the script is be run on files changed by a pull request, parse the PR number",
+    )
+
+    return parser.parse_args()
+
+
+def get_files_from_pr(pr_num):
+    """Return a list of changed files from a GitHub Pull Request
+
+    Arguments:
+        pr_num {str} -- Pull Request number to get modified files from
+
+    Returns:
+        {list} -- List of modified filenames
+    """
+    files = []
+    pr_url = f"https://api.github.com/repos/bids-standard/bids-specification/pulls/{pr_num}/files"
+    resp = requests.get(pr_url)
+
+    for item in resp.json():
+        files.append(item["filename"])
+
+    return files
+
+
+def filter_files(pr_num, start_phrase="src"):
+    """Filter modified files from a Pull Request by a start phrase
+
+    Arguments:
+        pr_num {str} -- Number of the Pull Request to get modified files from
+
+    Keyword Arguments:
+        start_phrase {str} -- Start phrase to filter changed files by
+                              (default: {"src"})
+
+    Returns:
+        {list} -- List of filenames that begin with the desired start phrase
+    """
+    files = get_files_from_pr(pr_num)
+    filtered_files = []
+
+    for filename in files:
+        if filename.startswith(start_phrase):
+            filtered_files.append(filename)
+
+    return filtered_files
+
+
+if __name__ == "__main__":
+    args = parse_args()
+    changed_files = filter_files(args.pull_request)
+    print(changed_files)

tools/requirements.txt

@@ -0,0 +1 @@
+requests