Hufflepdf/ChangeLog.md

# Revision history for Hufflepdf

## 0.2.0.1  -- 2019-11-27

* Fix bug discovered while running Hufflepdf on a PDF output from pdftk : magic keywords like `obj`, `stream` or `xref` can have spaces after them before the EOL

## 0.2.0.0  -- 2019-10-14

* Implement PDF's multilayer update mechanism

## 0.1.1.1  -- 2019-05-31

* Fix a bug preventing files with the «trailer» keyword on the same line as the dictionary that follows it to be parsed
* Improve error messages
* Support empty lines as comments

## 0.1.1.0  -- 2019-05-23

* Rewrite the parser using attoparsec instead of parsec (about 1.5x faster)
* Use strict ByteStrings instead of lazy ones for an additional gain in performance, simpler type interface (lazy ByteStrings are now used only for output because that's required by the ByteString Builder) and fewer import lines

## 0.1.0.0  -- 2019-05-18

* First version. Released on an unsuspecting world.
Finally opt for uppercase Hufflepdf and rename cabal package 2019-05-18 09:49:31 +02:00			`# Revision history for Hufflepdf`
Prototype successfully parsing (only last) startxref 2019-05-13 08:05:28 +02:00
Fix a bug preventing Hufflepdf from reading objects with a ' ' after the `obj` keyword 2019-11-27 18:01:03 +01:00			`## 0.2.0.1 -- 2019-11-27`

			* Fix bug discovered while running Hufflepdf on a PDF output from pdftk : magic keywords like `obj`, `stream` or `xref` can have spaces after them before the EOL

Release 0.2.0.0 2019-10-14 10:16:14 +02:00			`## 0.2.0.0 -- 2019-10-14`

			`* Implement PDF's multilayer update mechanism`

Release bugfix v0.1.1.1 2019-05-31 15:16:23 +02:00			`## 0.1.1.1 -- 2019-05-31`

			`* Fix a bug preventing files with the «trailer» keyword on the same line as the dictionary that follows it to be parsed`
			`* Improve error messages`
			`* Support empty lines as comments`

Go strict ByteStrings with attoparsec 2019-05-24 10:48:09 +02:00			`## 0.1.1.0 -- 2019-05-23`

			`* Rewrite the parser using attoparsec instead of parsec (about 1.5x faster)`
			`* Use strict ByteStrings instead of lazy ones for an additional gain in performance, simpler type interface (lazy ByteStrings are now used only for output because that's required by the ByteString Builder) and fewer import lines`

First useable version 2019-05-18 11:09:03 +02:00			`## 0.1.0.0 -- 2019-05-18`
Prototype successfully parsing (only last) startxref 2019-05-13 08:05:28 +02:00
			`* First version. Released on an unsuspecting world.`