Merge pull request #20 from castlabs/VTK-2434-striprtf-fonttbl-support

Vtk 2434 striprtf fonttbl support

Author: sannies

pycaption/pl_stt.py

@@ -1,3 +1,4 @@
+from striprtf import striprtf
 from .base import BaseReader, CaptionSet, CaptionList, Caption, CaptionNode
 from .exceptions import CaptionReadNoCaptions, InvalidInputError
 
@@ -39,6 +40,8 @@ def _parse_sub(self, sub):
         return sub_start, sub_end, sub_text
 
     def detect(self, content):
+        if content.startswith(u'{\\rtf1'):
+            content = striprtf.rtf_to_text(content)
         if self._get_header(content) and self._get_body(content):
             return True
         else:
@@ -59,6 +62,8 @@ def _guess_framerate(self, nonempty_splits):
     def read(self, content, lang="en-US"):
         if type(content) != str:
             raise InvalidInputError("The content is not a unicode string.")
+        if content.startswith(u'{\\rtf1'):
+            content = striprtf.rtf_to_text(content)
 
         try:
             header = self._get_header(content)

pycaption/stl/ebu_stl_reader.py

@@ -237,15 +237,19 @@ def decode(self, input, errors="strict"):
                 state = None
         return output, len(input)
 
-    def search(self, name):
-        if name in ("iso6937", "iso_6937-2"):
-            return codecs.CodecInfo(self.encode, self.decode, name="iso_6937-2")
+    def search(name):
+        if name in ("iso6937", "iso_6937-2", "iso_6937_2"):
+            return codecs.CodecInfo(
+                name="iso_6937-2",
+                encode=iso6937().encode,
+                decode=iso6937().decode,
+            )
 
     def encode(self, input, errors="strict"):
         pass
 
 
-codecs.register(iso6937().search)
+codecs.register(iso6937.search)
 
 
 class STLReader(BaseReader):

requirements.txt

@@ -5,4 +5,5 @@ Pillow>=7.0.0
 fonttools~=4.33.3
 langcodes~=3.3.0
 arabic-reshaper==3.0.0
-python-bidi==0.4.2
+python-bidi==0.4.2
+git+https://github.com/castlabs/striprtf.git@4d2a1fb9b1437fe068bdc3ffb1c87f0a4fe5d373#egg=striprtf

setup.py

@@ -7,7 +7,10 @@
     'README.rst')
 with open('requirements.txt') as f:
     requirements = f.read().splitlines()
+install_requires = [req for req in requirements if not req.startswith('git+')]
+install_requires += [req.split('#egg=')[1] + ' @ ' + req for req in requirements if req.startswith('git+')]
 
+print(install_requires)
 
 setup(
     name='pycaption',
@@ -17,7 +20,7 @@
     author='Sebastian Annies',
     author_email='[email protected]',
     url='https://github.com/castlabs/pycaption',
-    install_requires=requirements,
+    install_requires=install_requires,
     packages=find_packages(),
     include_package_data=True,
     classifiers=[

tests/ebu1991/test.stl

@@ -0,0 +1,15 @@
+import unittest
+from os.path import join, dirname, abspath
+
+from pycaption.stl.ebu_stl_reader import STLReader
+
+
+class SRTtoWebVTTTestCase(unittest.TestCase):
+    directory = join(dirname(abspath(__file__)))
+    def test_stl(self):
+        stl = STLReader()
+        content = open(join(self.directory, "ebu1991", "test.stl"), "rb").read()
+        self.assertTrue(stl.detect(content))
+        captions = stl.read(content)
+
+        self.assertEqual(len(captions.get_captions("de")), 494)