Update python-novice-inflammation (and split PO files)

helpers/splitpot.py

@@ -0,0 +1,172 @@
+from argparse import ArgumentParser
+import configparser
+import datetime
+import glob
+from pathlib import Path
+
+# from txclib.commands import _set_source_file
+
+# Input the lesson file that we want to split: splitplot.py lessonX.pot
+
+# Create directory with the file name: lessonX.pot => lessonX/
+
+# Save header in memory; read till first empty line
+
+# Read first line, pattern: -> extract filename function
+# #: lessonX/filename.md:YY
+# if '#:' in line[:2]
+#   read till `:` => line[:2]
+# Add to the filename with a number, for later join them properly together. -> save as 01_[folder-name_]file.pot
+# Keep searching for whitespaces, look next line, same filename? add to it.
+
+def extract_filename(line):
+    '''
+    provides the filename of a file from the po sections, like:
+
+    "#: /path/to/file/with/courious.name:55"
+
+    parameters
+    ----------
+    line: a string containing a line with a filename.
+
+    returns
+    -------
+    filename: the name of the file.
+    '''
+    if '#:' not in line[:2]:
+        raise TypeError("The current line is not one that contains a filename.")
+    line_bits = line.split(':')
+    filename = line_bits[1].strip().split('/')[-1]
+    return filename
+
+
+
+class Pofiles:
+    def __init__(self, filename):
+        self.filename = self._check_filename(filename)
+        self.lesson = self.filename.split('/')[-1].split('.')[0]
+
+    def _check_filename(self, filename):
+        # filename is a pot/po file?
+        if '.po' not in filename[-4:]:
+            raise TypeError(f"The file {filename} doesn't has the right extension (.pot/.po)")
+        try:
+            with open(filename, 'r') as file_handler:
+                self.pot_content = file_handler.readlines()
+            # TODO Check file follows the pot format standard
+        except:
+            raise TypeError("There's been some error reading the file")
+        return filename
+
+    def split(self, directory='transifex'):
+        # Check the given directory exists, if not create it:
+        split_directory = Path(directory)
+        if not split_directory.is_dir():
+            split_directory.mkdir(parents=True)
+        # Create a directory for this lessons
+        lesson_directory = split_directory / self.lesson / 'pot'
+        lesson_directory.mkdir(parents=True)
+        first_blank = self.pot_content.index('\n')
+        self.header = self.pot_content[:first_blank + 1]
+        all_files = {} # To contain filename: [content]
+        rest_file = self.pot_content[first_blank + 1:]
+        file_sect = None
+
+        for line in rest_file:
+            if not file_sect:
+                file_sect = extract_filename(line)
+            if file_sect not in all_files.keys():
+                all_files[file_sect] = self.header + [line]
+            else:
+                all_files[file_sect].append(line)
+                if line == '\n':
+                    file_sect = None
+
+        # Write dictionary into files
+        for order_file, (file_sect, content_sect) in enumerate(all_files.items()):
+            with open(lesson_directory / f"{order_file:02d}__{file_sect}.pot", 'w') as section:
+                section.writelines(content_sect)
+
+        # Generate transifex config file
+        tx_dir = lesson_directory.parent / '.tx'
+        tx_dir.mkdir()
+        config = configparser.ConfigParser()
+        config['main'] = {"host": "https://www.transifex.com"}
+        with open(tx_dir / 'config', 'w') as txconf:
+            config.write(txconf)
+        # TODO generate them directly
+        # for pot in pots:
+        #     _set_source_file(tx_dir, self.lesson, 'en', pot)
+
+
+
+    def join(self, source_dir, language):
+        # TODO check that filenames are there to be joint
+        # get all the files on source_dir/*.language.po
+        list_translations = glob.glob(f"{source_dir}/{language}/*.po")
+        list_translations.sort()
+        # read them all and join them with a single header # NOTE should we join the header info too? (eg., authors)
+        all_content = []
+        translators = []
+        last_touch = ('name', '"PO-Revision-Date: 2000-01-01 00:00:00+0000\\n"\n')
+        for file_translated in list_translations:
+            with open(file_translated, 'r') as section:
+                lines = section.readlines()
+            header = lines[:lines.index('\n') + 1]  # we can use the last header as a template.
+            porev_line = list(filter(lambda x: 'Revision-Date' in x, header))[0]
+            potrans_line = list(filter(lambda x: 'Last' in x, header))[0]
+            touch = self._extract_date_po(porev_line)
+            if touch > self._extract_date_po(last_touch[1]):
+                last_touch = (potrans_line, porev_line)
+            lines = lines[lines.index('\n') + 1:] + ['\n']
+            all_content.extend(lines)
+            translators_pos = header.index('# Translators:\n')
+            translators.extend(header[translators_pos + 1: header.index('# \n', translators_pos + 1)])
+
+        # find line with last revision date and replace it.
+        # NOTE Transifex is not updating that field!
+        header[header.index(porev_line)] = last_touch[1]
+        header[header.index(porev_line) + 1] = last_touch[0]
+
+        # get unique and sorted names of translators
+        translators = sorted(set(translators))
+        # FIXME add dots after each year for each translators
+        start = header.index("# Translators:\n")
+        end = header.index("# \n", start + 1)
+        header[start + 1: end] = translators
+
+        all_content = header +  all_content
+
+        # path from the original filename
+        path_filename = Path(self.filename).parent
+        # Write file next to original with the language key.
+        with open(path_filename / f"{self.lesson}.{language}.po", 'w') as full_translation:
+            full_translation.writelines(all_content)
+
+    def _extract_date_po(self, line):
+        title, date, time = line.split()
+        time = time.split('\\n"')[0]  # removes the inside and outside linebreak
+        seconds_exist = time.count(":") == 2
+        if not seconds_exist:
+            time, zone = time.split('+')
+            time = time + ":00"
+            time = "+".join([time, zone])
+
+        return datetime.datetime.strptime(f"{date} {time}", "%Y-%m-%d %H:%M:%S%z")
+
+if __name__ == "__main__":
+    command = ArgumentParser(description="Breaks a pot files into smaller chunks for better management for the translators.")
+    command.add_argument('filename', help="pot file to be split.")
+    command.add_argument('--join', '-j', dest='source', help="source directory from where to join split files")
+    command.add_argument('--lang', '-l', dest='language', help="which language you want to combine")
+    arguments = command.parse_args()
+
+    pof = Pofiles(arguments.filename)
+
+    if arguments.source and not arguments.language:
+        command.error("--join requires --lang")
+
+    if arguments.source and arguments.language:
+        pof.join(arguments.source, arguments.language)
+    else:
+        pof.split()