1 files changed, 72 insertions, 0 deletions
diff --git a/src/str_format.py b/src/str_format.py
new file mode 100644
index 0000000..5d8c412
--- /dev/null
+++ b/src/str_format.py
@@ -0,0 +1,72 @@
+import pathlib
+import re
+import unicodedata
+
+def safe_path(name):
+    simplified = ''.join([alnum_or_space(c) for c in unaccent(name.lower())])
+    return '-'.join(simplified.split())
+
+def unaccent(s):
+    return ''.join(c for c in unicodedata.normalize('NFD', s) if unicodedata.category(c) != 'Mn')
+
+def alnum_or_space(c):
+    if c.isalnum():
+        return c
+    else:
+        return ' '
+
+def cleanup_text(s, lang):
+    s = re.sub('\'', '’', s)
+    s = re.sub(r'\.\.\.', '…', s)
+    s = re.sub(r'\. \. \.', '…', s)
+    s = cleanup_double_quotes(s, lang)
+    s = cleanup_paragraphs(s)
+
+    if lang == 'fr':
+        s = re.sub('“', '«', s)
+        s = re.sub('”', '»', s)
+
+        # Replace space by insecable spaces
+        s = re.sub(r' ([:?\!])', r' \1', s)
+        s = re.sub('« ', '« ', s)
+        s = re.sub(' »', ' »', s)
+
+        # Add missing insecable spaces
+        s = re.sub(r'([^ ]):', r'\1 :', s)
+        s = re.sub(r'([^ ])\?', r'\1 ?', s)
+        s = re.sub(r'([^ ])\!', r'\1 !', s)
+        s = re.sub(r'([^ ])»', r'\1 »', s)
+        s = re.sub(r'«([^ ])', r'« \1', s)
+
+    elif lang == 'en':
+        s = re.sub('« ', '“', s)
+        s = re.sub(' »', '”', s)
+        s = re.sub('«', '“', s)
+        s = re.sub('»', '”', s)
+
+    return s
+
+def cleanup_double_quotes(s, lang):
+    res = ''
+    quoted = False
+    for c in s:
+        if c == '"':
+            if quoted:
+                quoted = False
+                if lang == 'fr':
+                    res += '»'
+                elif lang == 'en':
+                    res += '”'
+            else:
+                quoted = True
+                if lang == 'fr':
+                    res += '«'
+                elif lang == 'en':
+                    res += '“'
+        else:
+            res += c
+    return res
+
+def cleanup_paragraphs(s):
+    ps = [f'      {p.strip()}' for p in re.split(r'\n+', s) if p.strip()]
+    return '\n\n'.join(ps)