Pdfstream as cmap #283
Pdfstream as cmap #283
Conversation
| @@ -140,7 +142,13 @@ def do_keyword(self, pos, token): | |||
|
|
|||
|
|
|||
| NIBBLES = ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.', 'e', 'e-', None, '-') | |||
|
|
|||
| IDENTITY_ENCODER = {'Identity-H':'Identity-H', | |||
There was a problem hiding this comment.
You mentioned that DLIdent-* is not in the PDF reference manual. Did you find other pdf documentation that mentions this? If so, you could add a comment that refers to it such that we do not remove DLIdent-* by accident.
There was a problem hiding this comment.
I haven't seen any pdf with DLIdent-* but have included it due to its harmless nature. Will also include code comment to avoid DLIdent-* being removed by accident.
pdfminer/cmapdb.py
Outdated
| return struct.unpack('>%dB' % n, code) | ||
| else: | ||
| return () | ||
|
|
||
| ## UnicodeMap |
There was a problem hiding this comment.
There are a lot of these unnecessary code comments in cmapdb.py. Since you are editing this file anyway, could you remove those? And likewise for pdffont.py?
There was a problem hiding this comment.
Sure, Can do that.
pdfminer/pdffont.py
Outdated
| @@ -140,12 +130,18 @@ def do_keyword(self, pos, token): | |||
|
|
|||
|
|
|||
| NIBBLES = ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.', 'e', 'e-', None, '-') | |||
| """ | |||
There was a problem hiding this comment.
I might be nitpicking here: according to pep8 block comments should start with a hashtag instead of using a multi-line string.
Advantage of using hashtag: an intelligent editor can understand that it is a comment
|
@pietermarsman Hi, Have removed the comments. Can you please review it once again and merge if no other changes are required. :) |
|
Hi @fakabbir, sorry for the miscommunication but I do not have commit rights right now. I do my bit by reviewing and creating PR's. I hope to get commit rights soon though. |
There was a problem hiding this comment.
This looks like euske/pdfminer/pull/179 with some extra features and tests. It'd be great to have this merged asap.
Resolves #210 .
The previous work #264 maps "OneByteEncodingH/V" to default CMap initialization. It also ignores the fact that "DLIdent-H/V" is a valid encoding since it isn't mentioned in any PDF Reference the author can dig in.
While looking for the output of pdf2text.py the OneByteEncoding characters were absent denoting that default CMap encoding is ineffective to capture the glyph.
I have added the decode function for OneByteEncoding and also added DLIdent as there is a possibility of this being in many PDF.
Both OneByteEncoding and DLIdent aren't available in any PDF Reference referred.