|
|
23186100a8
|
Reimplement getObj with the newest tools in PDF.Object.Navigation, in particular implement browsing by paths or random objectId access
|
2020-02-15 10:25:09 +01:00 |
|
|
|
b916ab5206
|
Just noticed Streams are a kind of Dictionary too, since they have a header
|
2020-02-15 10:23:32 +01:00 |
|
|
|
4a6dbda7d3
|
Move Error type from Pages to Navigation as a candidate for MonadFail required by PDFContent defined there
|
2020-02-15 10:22:42 +01:00 |
|
|
|
923d1800b0
|
Gain a bit of speed by using native Attoparsec for number types instead of reimplementing them with ByteString conversion and call to read
|
2020-02-14 18:02:40 +01:00 |
|
|
|
1c457d71d8
|
Fix the reading of Hexadecimal string objects detected by running the tests implemented from the spec
|
2020-02-14 18:00:12 +01:00 |
|
|
|
a72d76e229
|
Add unit tests to make sure I'm not breaking things too much
|
2020-02-14 17:58:03 +01:00 |
|
|
|
919f640443
|
Merge branch 'extract-text' into navigation
|
2020-02-12 17:35:56 +01:00 |
|
|
|
32f9866106
|
Use peek to improve directObject parser avoiding a large <|> disjunction
|
2020-02-12 17:34:27 +01:00 |
|
|
|
eb4d76002c
|
Finish the split of Navigation out of Page, generalize the use of MonadFail with a custom Error monad (~= Either String)
|
2020-02-11 22:41:46 +01:00 |
|
|
|
af994cb50c
|
WIP: in the process of migrating to Object.Navigation in Pages, still unsure how to manage simple Content parsing and efficient font loading (+ giving a way to edit Contents)
|
2020-02-11 17:59:15 +01:00 |
|
|
|
704d7a7fcf
|
It turns out Output.concat wasn't necessary, OBuilder seems already is a Monoid so mconcat works (that fact was used in the very implementation of concat…)
|
2020-02-11 17:36:29 +01:00 |
|
|
|
11647eb4eb
|
Implement output for Content streams
|
2020-02-11 17:26:47 +01:00 |
|
|
|
aed7af376a
|
WIP: still trying to figure things out, moved to a separate submodule for Navigation, proper naming is hell
|
2020-02-11 08:29:08 +01:00 |
|
|
|
e77bbbcda9
|
WIP: start moving some navigation-related routines from Pages into Object directly and generalize them to multi-component to allow easier browsing
|
2020-02-10 17:43:04 +01:00 |
|
|
|
195446e653
|
Allow resources with no /Font field, they won't cause any problem as long as no call to Tf (to load a font) is made
|
2020-02-10 17:41:44 +01:00 |
|
|
|
9f1b1afafe
|
Implement Text rendering from parsed Content
|
2020-02-10 10:54:44 +01:00 |
|
|
|
20466c4f13
|
WIP: Clean code parsing «pages» (now Content), separated from text rendering (will be reimplemented as an upper layer, also providing modification as stream filters) — Page is also forgotten for now, will need a big improvement in Object navigation
|
2020-02-09 22:42:57 +01:00 |
|
|
|
325250383a
|
Add support for fonts and implement MacRomanEncoding
|
2020-02-08 08:15:32 +01:00 |
|
|
|
c48ab22808
|
Forgot some useless parentheses when playing with operator precedences
|
2020-02-04 17:05:15 +01:00 |
|
|
|
a2b66ac6d6
|
Generalize the getFont function because some /Resources have a direct dictionary as value for their /Font property
|
2020-02-04 17:04:42 +01:00 |
|
|
|
cefb08ee50
|
Going a step further in «optimization» (slowing it even more…) by replacing choice by a search in a Map
|
2019-11-30 21:46:22 +01:00 |
|
|
|
afbbcbffc5
|
Finish implementing the new stack-based call parser
|
2019-11-30 12:39:40 +01:00 |
|
|
|
bac08446dd
|
WIP: starting to fix this criminally inefficient parser for PDF's postfix-operator instructions
|
2019-11-29 17:42:57 +01:00 |
|
|
|
f9f799c59b
|
Take the dirty code of «getText» and turn it into a relatively clean module exposing pages, that can be retrieved all at once or by page number (numbered human-style, starting from 1)
|
2019-11-29 11:51:35 +01:00 |
|
|
|
08a9717b3a
|
Get rid of wrapper PageContents structure returned by PageContent in the PDF.Text module (and return directly [ByteString] instead)
|
2019-11-29 11:48:28 +01:00 |
|
|
|
42a02808c1
|
Merge branch 'main' into extract-text
|
2019-11-27 18:05:47 +01:00 |
|
|
|
380c1e439b
|
Fix a bug preventing Hufflepdf from reading objects with a ' ' after the obj keyword
|
2019-11-27 18:01:19 +01:00 |
|
|
|
c9f050e64b
|
Remove deprecated debug script and forgotten comments to bypass the selective export of Text module
|
2019-10-14 10:17:15 +02:00 |
|
|
|
3a3e1533b4
|
Clean ByteString types to identify when a ByteString contains the representation of an integer in a given base and fix the last remaining PDF string (un)escaping issue
|
2019-10-14 10:17:15 +02:00 |
|
|
|
a96e36ec5a
|
Fix error silently discarding code ranges, make sure ByteString intervals are created with the correct byte length and decode utf16BE encoded values in single-value ranges
|
2019-10-14 10:17:15 +02:00 |
|
|
|
d07c286f8e
|
Clean exported ByteString custom functions
|
2019-10-14 10:17:15 +02:00 |
|
|
|
7a15113285
|
Try and re-implement string decoding — compiles but now fails to decode any string
|
2019-10-14 10:17:15 +02:00 |
|
|
|
36d7f9b819
|
Still debugging, broke pretty much everything and finally implementing a proper coderange parsing for CMap because apparently that's necessary
|
2019-10-14 10:17:15 +02:00 |
|
|
|
b8ca7281aa
|
Fix parsing errors forgetting to make sure there's a space after special operator arguments like names and stringObjects
|
2019-10-14 10:17:15 +02:00 |
|
|
|
32efdcdd6b
|
Try and fix stuff by generalizing a signature to ease debugging and add parenthesis which I think should have been here all along
|
2019-10-14 10:17:15 +02:00 |
|
|
|
3b59fd0c61
|
Separate CMap and Text in two distinct modules
|
2019-10-14 10:17:15 +02:00 |
|
|
|
0374b72920
|
Finish implementing reading, still bugs to investigate
|
2019-10-14 10:17:15 +02:00 |
|
|
|
1dd22c3889
|
Going to try with Text, naturally handling UTF-16 but will still have to parse «int codes» manually from strings
|
2019-10-14 10:17:15 +02:00 |
|
|
|
98d029c4d4
|
In complete debug, more or less implemented CMap parsing but apparently it uses UTF16 ?!
|
2019-10-14 10:17:15 +02:00 |
|
|
|
c349d9b4c2
|
Don't trust serializer, they have nothing todo with a reasonable binary encoding
|
2019-10-14 10:17:15 +02:00 |
|
|
|
e7484ef536
|
Completely lost, the same old Char8 / Word8 again, implemented all the text reading, still needing a couple details to parse CMaps
|
2019-10-14 10:17:15 +02:00 |
|
|
|
f9e5683bf4
|
WIP: Use previous changes to start implementing font caching and text parsing (still very broken, doesn't compile)
|
2019-10-14 10:17:15 +02:00 |
|
|
|
b8eb9e6856
|
Generalize the Parser type into a MonadParser class to use with MonadTrans and remove redundant code already defined in Applicative or Attoparsec
|
2019-10-14 10:17:15 +02:00 |
|
|
|
66d315b7fe
|
Reflect the distinction between eval and run from State monad into the Parser module
|
2019-10-14 10:17:15 +02:00 |
|
|
|
51db57ec67
|
Ugly commit, breaks everything, still trying to figure a grammar for text
|
2019-10-14 10:17:15 +02:00 |
|
|
|
6f3c159ea7
|
Adding a module to implement text reading and a demo program to go with it
|
2019-10-14 10:17:15 +02:00 |
|
|
|
68f90d20e2
|
Implement PDF's multilayer updates and use it in getObj to display only the current version of the object taken into account instead of the concatenation of all its versions
|
2019-09-22 01:40:39 +02:00 |
|
|
|
3a39c75e6a
|
Stop requiring an empty line between subsections in a xref section
|
2019-09-22 01:37:28 +02:00 |
|
|
|
29c5823f34
|
Fix precision bug caused by using Floats to represent PDF Number values sometimes used to represent a byte offset within a file
|
2019-09-22 01:34:17 +02:00 |
|
|
|
699f830a45
|
Simplify XRef structure, clarify integer types and remove nextLine
|
2019-09-20 22:39:14 +02:00 |
|
