From 4dec09c7952a6ddc186d7682db129033ada67f38 Mon Sep 17 00:00:00 2001
From: Roland Hieber <rohieb@rohieb.name>
Date: Wed, 25 Dec 2013 21:47:43 +0100
Subject: [PATCH] projects: pdf highlighter

---
 blag/post/pdf-highlighter.mdwn | 18 ++++++++++
 projects.mdwn                  |  3 ++
 projects/pdf.lang              | 63 ++++++++++++++++++++++++++++++++++
 3 files changed, 84 insertions(+)
 create mode 100644 blag/post/pdf-highlighter.mdwn
 create mode 100644 projects/pdf.lang

diff --git a/blag/post/pdf-highlighter.mdwn b/blag/post/pdf-highlighter.mdwn
new file mode 100644
index 0000000..4d0aad8
--- /dev/null
+++ b/blag/post/pdf-highlighter.mdwn
@@ -0,0 +1,18 @@
+[[!meta title="A Highlighting plugin for PDF files"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+[[!img defaults size=x200]]
+
+In my [[last post|optimizing-xsane-s-scanned-pdfs]], I used ikiwiki‘s [highlight
+plugin][iwhl] to highlight PDF files. But since the underlying [highlight][]
+library did not support highlighting for PDF files yet, I had to write my own
+highlighting definition. Due to limitiations in the library, it's not perfect;
+for example, it does not highlight things inside streams, but in case you’re
+interested, you can get the source:
+
+* [`pdf.lang`](/projects/pdf.lang)
+
+[iwhl]: http://ikiwiki.info/plugins/highlight/
+[highlight]: http://www.andre-simon.de/doku/highlight/en/highlight.php
+
+[[!tag PDF meta project highlight]]
diff --git a/projects.mdwn b/projects.mdwn
index 4a54865..8d22ce4 100644
--- a/projects.mdwn
+++ b/projects.mdwn
@@ -7,6 +7,9 @@ of them finished.
 
 * [[projects/geojackal]] – A geocache management database
 * [[projects/infopoint-html]] – A full screen web page viewer for info terminals
+* [`pdf.lang`](pdf.lang) – PDF syntax description for the [highlight][] library
+
+[highlight]: http://www.andre-simon.de/doku/highlight/en/highlight.php
 
 ## Ideas for new software projects
 
diff --git a/projects/pdf.lang b/projects/pdf.lang
new file mode 100644
index 0000000..7ed6f64
--- /dev/null
+++ b/projects/pdf.lang
@@ -0,0 +1,63 @@
+-- vim: set ft=lua ts=2 sw=2 et :
+-- Language definition for PDF files
+-- Author: Roland Hieber <rohieb@rohieb.name>
+-- Date: 2013-11-22
+-- Known Bugs: 
+--  * Does not highlight files with MacOS (CR-only) line endings
+--  * Does not (yet) highlight inside streams due to limitations in the library
+--
+
+Description="Highlighting definitions for the Portable Document Format (PDF)"
+
+IgnoreCase=false
+
+-- File Structure
+PreProcessor={
+  Prefix=[[%PDF-[0-9]\.[0-9]|%%EOF|xref|startxref|trailer]]
+}
+
+-- Comments. But do not match file structure elements.
+Comments={
+  { Block=false,
+    Delimiter={ [[%(?!PDF-[0-9]\.[0-9]|%EOF)]] },
+  },
+}
+
+-- Numbers: 0.45, +1.34, -.4, 123, 4., and so on.
+Digits=[[ [-+]?\.[0-9]+|[-+]?[0-9]+\.?[0-9]* ]]
+
+-- Strings: (string), <hex> and streams
+Strings={
+  DelimiterPairs= {
+    { Open=[[ \( ]], Close=[[ \) ]] },
+    { Open=[[ < ]], Close=[[ > ]] },
+    { Open=[[ ^stream ]], Close=[[ ^endstream ]], Raw=true },
+  }
+}
+
+-- Note: we highlight dictionary and array syntax as "keywords", so we have to
+-- include them in Identifiers. This definition basically matches the allowed
+-- characters for Names. Also, we do not want to match Numbers, Streams,
+-- References and file structure elements as identifiers
+Identifiers=[[ (?!%PDF-[0-9]\.[0-9]|%%EOF|xref|startxref|trailer|[0-9]+\s+[0-9]+\s+(R|obj)|[-+]?\.[0-9]+|[-+]?[0-9]+\.?[0-9]*)[^\s\[\]\(\){}<>/%]+ ]]
+
+Keywords={
+  -- Indirect Objects
+  { Id=1,
+    Regex=[[ [0-9]+\s+[0-9]+\s+(obj|R)|endobj]],
+    Group=0
+  },
+  -- Arrays and Dictionaries
+  { Id=2,
+    Regex=[[ \[|\]|<<|>> ]],
+  },
+  -- Names
+  { Id=3,
+    Regex=[[ /[^\s\[\]\(\){}<>/%]+ ]],
+  },
+  -- Constants
+  { Id=4,
+    --List={"true", "false", "null"},
+    Regex=[[ true|false|null ]],
+  },
+}
-- 
2.20.1