Menu
▾
▴
[Oorexx-svn] SF.net SVN: oorexx:[4912] incubator/DocMusings
|
From: <jfa...@us...> - 2009-07-13 01:47:31
|
Revision: 4912
http://oorexx.svn.sourceforge.net/oorexx/?rev=4912&view=rev
Author: jfaucher
Date: 2009-07-13 01:47:23 +0000 (Mon, 13 Jul 2009)
Log Message:
-----------
Work in progress, 1st step reached.
The XML parser is operational on the oorexx doc files.
I can read a SGML file and print it on stdout unchanged (same layout).
Tested on fundclasses.sgml
Next step : implement the option -fix which will let generate a valid XML.
Modified Paths:
--------------
incubator/DocMusings/makevalidxml/_diary.txt
incubator/DocMusings/makevalidxml/makevalidxml.rex
incubator/DocMusings/makevalidxml/myxmlparser.cls
incubator/DocMusings/railroad/_diary.txt
incubator/DocMusings/railroad/generate.rex
incubator/DocMusings/railroad/syntaxdiagram2svg/css/syntaxdiagram.css
incubator/DocMusings/railroad/syntaxdiagram2svg_out/DO-conditional.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/DO-repetitor.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/DO.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/Decimal integer.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/Hexadecimal literal.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/IPV4 address in dot form-IPv4-range integer.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/IPV4 address in dot form-decimal digit.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/IPV4 address in dot form.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/MathLoadFuncs.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/RxCalcSqrt.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/URL syntax 1.png
incubator/DocMusings/railroad/syntaxdiagram2svg_out/URL syntax 2.png
Modified: incubator/DocMusings/makevalidxml/_diary.txt
===================================================================
--- incubator/DocMusings/makevalidxml/_diary.txt 2009-07-13 00:37:08 UTC (rev 4911)
+++ incubator/DocMusings/makevalidxml/_diary.txt 2009-07-13 01:47:23 UTC (rev 4912)
@@ -1,4 +1,36 @@
===============================================================================
+2009 july 12
+
+The first step has been reached...
+The command makevalidxml can read a SGML file and print it to stdout unchanged (same layout).
+But behind the scene, I have all the informations needed to transform the XML where needed.
+The following options remain to implement :
+-fix
+-xinclude
+
+
+===============================================================================
+2009 july 11
+
+I need effectively to modify the XML parser, to keep the text as-is :
+- getchar returns EOL as any other char.
+- Each chunk has a value in its text property, to be used for printing as-is.
+
+
+Things to fix to have a valid XML :
+
+<?xml version="1.0" standalone="no">
+must be
+<?xml version="1.0" standalone="no"?>
+
+<!DOCTYPE article PUBLIC "-//OASIS//DTD DocBook V4.2//EN"
+must have a system identifier
+<!DOCTYPE article PUBLIC "-//OASIS//DTD DocBook V4.2//EN" "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd"
+
+<imagedata> must be completed by </imagedata>
+
+
+===============================================================================
2009 june 27
First step :
Modified: incubator/DocMusings/makevalidxml/makevalidxml.rex
===================================================================
--- incubator/DocMusings/makevalidxml/makevalidxml.rex 2009-07-13 00:37:08 UTC (rev 4911)
+++ incubator/DocMusings/makevalidxml/makevalidxml.rex 2009-07-13 01:47:23 UTC (rev 4912)
@@ -39,53 +39,277 @@
/* */
/*----------------------------------------------------------------------------*/
+/****
+Usage :
+ makevalidxml [-debug] [-dump] [-fix] [-help] [-xinclude] <filename>
+Description :
+ This script is intended to work on the XML files of the ooRexx doc.
+ It reads the file <filename>, parses it and sends it to stdout.
+ By default, the layout of the output file is kept the most similar possible
+ to the layout of the input.
+ Options :
+ -debug : Insert additional informations in the ouptut.
+ -dump : The elements and attributes are dumped without attempting to
+ keep the original layout.
+ -fix : The output is modified where needed, to make it valid XML.
+ -xinclude : Use <xi:include> instead of XML entities to include files.
+****/
--- JLF : Nothing started, yet...
-use strict arg filename
+parse source . callType .
+arguments = .Arguments~new(callType, arg(1,"array"))
+if arguments~error <> "" then do
+ call Help arguments~error
+ return 1
+end
+
+if arguments~helpOption then do
+ call Help
+ return 0
+end
+
parser = .myparser~new()
-errortxt = parser~parse_file(filename)
+parser~debug = arguments~debugOption
+parser~dump = arguments~dumpOption
+errortxt = parser~parse_file(arguments~filename)
+if errortxt <> "" then do
+ say errortxt
+ return 1
+end
-return
+return 0
::requires 'myxmlparser.cls'
::class myparser subclass xmlparser
+::attribute dump
+
::method start_element
-use arg chunk
-call charout , '<'chunk~tag
-if chunk~attr <> .nil then do f over chunk~attr
- call charout , ' 'f'="'self~textxlate(chunk~attr[f])'"'
- end
-say '>'
-return
+ use arg chunk
+ if self~dump then do
+ if self~debug then say '[start_element:'chunk~line':'chunk~col'['
+ call charout , '<'chunk~tag
+ if chunk~attr <> .nil then
+ do f over chunk~attr
+ call charout , ' 'f'="'self~textxlate(chunk~attr[f])'"'
+ end
+ call charout , '>' -- say '>'
+ if self~debug then say ']start_element]'
+ end
+ else do
+ if self~debug then call charout , '[start_element:'chunk~line':'chunk~col'['
+ call charout , '<'chunk~text'>'
+ if self~debug then call charout , ']start_element]'
+ end
+ return
::method end_element
-use arg chunk
-say '</'chunk~tag'>'
-return
+ use arg chunk
+ if self~dump then do
+ if self~debug then say '[end_element'chunk~line':'chunk~col'['
+ call charout , '</'chunk~tag'>' -- say '</'chunk~tag'>'
+ if self~debug then say ']end_element]'
+ end
+ else do
+ if self~debug then call charout , '[end_element:'chunk~line':'chunk~col'['
+ if chunk~text = '' then call charout , '</'chunk~tag'>'
+ if self~debug then call charout , ']end_element]'
+ end
+ return
::method passthrough
-use arg chunk
-say '<'chunk~text'>'
-return
+ use arg chunk
+ if self~dump then do
+ if self~debug then say '[passthrough:'chunk~line':'chunk~col'['
+ call charout , '<'chunk~text'>' -- say '<'chunk~text'>'
+ if self~debug then say ']passthrough]'
+ end
+ else do
+ if self~debug then call charout , '[passthrough:'chunk~line':'chunk~col'['
+ call charout , '<'chunk~text'>'
+ if self~debug then call charout , ']passthrough]'
+ end
+ return
::method text
-use arg chunk
-say chunk~text
-return
+ use arg chunk
+ if self~dump then do
+ if self~debug then call charout , '[text:'chunk~line':'chunk~col'['
+ call charout , chunk~text
+ if self~debug then say ']text]'
+ end
+ else do
+ if self~debug then call charout , '[text:'chunk~line':'chunk~col'['
+ call charout , chunk~text
+ if self~debug then call charout , ']text]'
+ end
+ return
::method error
-use arg err
-say err~text
-return
+ use arg err
+ say err~text
+ return
-::method textxlate private
-use arg text
-text = text~changestr('&', '&') -- always do this one first!
-text = text~changestr('>', '>')
-text = text~changestr('<', '<')
-return text
+::method xlatetext private
+ use arg text
+ /*
+ text = text~changestr('&', '&') -- always do this one first!
+ text = text~changestr('>', '>')
+ text = text~changestr('<', '<')
+ */
+ return text
+-------------------------------------------------------------------------------
+::class Arguments
+::attribute debugOption
+::attribute dumpOption
+::attribute error
+::attribute filename
+::attribute fixOption
+::attribute helpOption
+::attribute xincludeOption
+
+::method init
+ self~debugOption = .false
+ self~dumpOption = .false
+ self~error = ""
+ self~filename = ""
+ self~fixOption = .false
+ self~helpOption = .false
+ self~xincludeOption = .false
+
+ -- Tokenize the arguments, if needed
+ use strict arg callType, arguments -- always an array
+ select
+ when callType == "COMMAND" & arguments~items == 1 then args = String2Args(arguments[1])
+ when callType == "SUBROUTINE" & arguments~items == 1 & arguments[1]~isA(.array) then args = arguments[1]
+ otherwise args = arguments
+ end
+
+ -- Use makeArray to have a non-sparse array,
+ -- because omitted parameters have no corresponding index,
+ -- and we ignore omitted parameters here.
+ args = args~makeArray
+
+ -- Process the options
+ loop i=1 to args~items
+ option = args[i]
+ if option~left(1) <> "-" then leave
+ select
+ when "-debug"~caseLessEquals(option) then self~debugOption = .true
+ when "-dump"~caseLessEquals(option) then self~dumpOption = .true
+ when "-fix"~caseLessEquals(option) then self~fixOption = .true
+ when "-help"~caseLessEquals(option) then self~helpOption = .true
+ when "-xinclude"~caseLessEquals(option) then self~xincludeOption = .true
+ otherwise do
+ self~error = "Unknown option" option
+ return
+ end
+ end
+ -- Return now if help requested
+ if self~helpOption then return
+ end
+
+ -- Process the arguments
+ if i > args~items then do
+ self~error = "<filename> is missing"
+ return
+ end
+ self~filename = args[i]~strip
+ i += 1
+ if i > args~items then return
+ self~error = "Unexpected arguments :" args~section(i)~toString("L", " ")
+ return
+
+
+-------------------------------------------------------------------------------
+::routine String2Args public
+ -- Converts a string to an array of arguments.
+ -- Arguments are separated by whitespaces (anything < 32) and can be quoted.
+ -- This routine tries to follow the behavior of cmd.exe, which lets write such things :
+ -- %1 %2 %3
+ -- myscript c:\dir1\my dir\dir 2 c:\dir1\my dir\dir 2
+ -- myscript "c:\dir1\my dir\dir 2" "c:\dir1\my dir\dir 2"
+ -- myscript c:\dir1\"my dir"\dir 2 c:\dir1\"my dir"\dir 2
+ -- myscript c:\dir1\my dir\dir" 2" c:\dir1\my dir\dir" 2"
+ -- myscript he says "I told you "hello"!" he says "I told you "hello"!"
+ -- myscript he says "I told you ""hello""!" he says "I told you ""hello""!"
+ --
+ -- Unlike cmd, this routine does an additional processing :
+ -- Quotes are removed from the value, except when they are doubled inside a quoted string.
+ -- cmd parameter value in args
+ -- c:\dir1\my 'c:\dir1\my'
+ -- "c:\dir1\my dir\dir 2" 'c:\dir1\my dir\dir 2'
+ -- c:\dir1\"my dir"\dir 'c:\dir1\my dir\dir'
+ -- "I told you "hello"!" 'I told you hello!'
+ -- "I told you ""hello""!" 'I told you "hello"!'
+ use strict arg string
+
+ args = .Array~new
+ i = 1
+
+ loop label arguments
+ -- Skip whitespaces
+ loop
+ if i > string~length then return args
+ if string~subchar(i) > " " then leave
+ i += 1
+ end
+
+ -- Process current argument : can be made of several chunks, all chunks are concatenated
+ -- Example chunk1 chunk2 chunk3 value
+ -- one one 'one'
+ -- "one two" "one two" 'one two'
+ -- one" two "three one " two " three 'one two three'
+ -- one" ""two"" "three one " ""two"" " three 'one "two" three'
+ current = .MutableBuffer~new
+ loop label current_argument
+ if string~subchar(i) == '"' then do
+ -- Chunk surrounded by quotes : whitespaces are kept, double occurrence of quotes are replaced by a single embedded quote
+ loop label quoted_chunk
+ i += 1
+ if i > string~length then return args~~append(current~string)
+ select
+ when string~subchar(i) == '"' & string~subchar(i+1) == '"' then do
+ current~append('"')
+ i += 1
+ end
+ when string~subchar(i) == '"' then do
+ i += 1
+ leave quoted_chunk
+ end
+ otherwise current~append(string~subchar(i))
+ end
+ end quoted_chunk
+ end
+ if string~subchar(i) <= " " then do
+ args~append(current~string)
+ leave current_argument
+ end
+ -- Chunk not surrounded by quotes : ends when a whitespace or quote is reached
+ loop
+ if i > string~length then return args~~append(current~string)
+ if string~subchar(i) <= " " | string~subchar(i) == '"' then leave
+ current~append(string~subchar(i))
+ i += 1
+ end
+ end current_argument
+ end arguments
+ return args
+
+-------------------------------------------------------------------------------
+::routine Help
+ use strict arg errorMessage=""
+ if errorMessage <> "" then say errorMessage
+ -- The help text is taken from the comment at the begining of the source file.
+ display = .false
+ do line = 1 to sourceline()
+ sourceline = sourceline(line)
+ if sourceline~pos("****/") <> 0 then leave
+ if display then say sourceline
+ if sourceline~pos("/****") <> 0 then display = .true
+ end
+ return
+
Modified: incubator/DocMusings/makevalidxml/myxmlparser.cls
===================================================================
--- incubator/DocMusings/makevalidxml/myxmlparser.cls 2009-07-13 00:37:08 UTC (rev 4911)
+++ incubator/DocMusings/makevalidxml/myxmlparser.cls 2009-07-13 01:47:23 UTC (rev 4912)
@@ -84,8 +84,7 @@
JLF : adaptation of xmlparser.cls to my own needs.
- reprint the parsed XML the most possible like the original one.
- add missing closing tags
-Don't know yet if can do that only by overriding the public methods, so I made
-a copy of the XML parser.
+I need to modify the parser, to keep the text unchanged, hence this copy...
*/
@@ -112,8 +111,14 @@
::method charidx attribute private -- the index into a xml line
::method errortxt attribute private -- error text
::method eof attribute private -- done parsing?
+::attribute endofline -- derived from the first line of the xml file, assume all lines have this EOL
+::attribute lastlineendofline -- except the last line which may have no EOL
+::attribute debug
+::method init
+self~debug = .false
+
/*----------------------------------------------------------------------------*/
/* Method: create_error */
/* Description: creates an xmlerror instance. */
@@ -143,15 +148,49 @@
/*----------------------------------------------------------------------------*/
+/* Method: detect_eol_style */
+/* Description: detect the EOL style of the first line and last line */
+/*----------------------------------------------------------------------------*/
+::method detect_eol_style
+use strict arg stream
+self~endofline = ""
+previouschar = ""
+stream~seek("=1")
+do forever
+ currentchar = stream~charin
+ if currentchar == "" then leave -- EOF
+ if c2x(currentchar) == "0A" then do
+ if c2x(previouschar) == "0D" then self~endofline = x2c("0D0A")
+ else self~endofline = x2c("0A")
+ leave
+ end
+ previouschar = currentchar
+end
+
+-- It's not possible to make a distinction between a last (non empty) line without EOL
+-- and a last (non empty) line with EOL, when working with the array returned by stream~arrayIn.
+-- Hence this attribute...
+self~lastlineendofline = ""
+stream~seek("<1")
+lastchar = stream~charin
+if c2x(lastchar) == "0A" then self~lastlineendofline = self~endofline
+return
+
+
+/*----------------------------------------------------------------------------*/
/* Method: current char */
-/* Description: return the current character. */
+/* Description: return the current character, or one of the relative */
+/* characters of the current line */
/*----------------------------------------------------------------------------*/
::method currentchar private
expose src lineidx charidx
-use strict arg
-if lineidx > src~items then return .nil
-return src[lineidx]~substr(charidx, 1)
+use strict arg offset=0 -- can be negative
+if lineidx > src~items then return ''
+charpos = charidx + offset
+if charpos < 1 then return ''
+if charpos > src[lineidx]~length then return ''
+return src[lineidx]~substr(charpos, 1)
/*----------------------------------------------------------------------------*/
@@ -162,16 +201,22 @@
::method getchar private
expose src lineidx charidx eof
use strict arg
-character = src[lineidx]~substr(charidx, 1)
-charidx = charidx + 1
+if lineidx > src~items then do
+ eof = .true
+ return ''
+ end
if charidx > src[lineidx]~length then do
lineidx = lineidx + 1
charidx = 1
+ if lineidx > src~items then do
+ if self~lastlineendofline <> "" then return self~lastlineendofline
+ eof = .true
+ return ''
+ end
+ return self~endofline
end
-if lineidx > src~items then do
- eof = .true
- return ''
- end
+character = src[lineidx]~substr(charidx, 1)
+charidx = charidx + 1
return character
@@ -186,33 +231,39 @@
errlineidx = lineidx
errcharidx = charidx
chunk = .xmlchunk~new
+chunk~line = lineidx
+chunk~col = charidx
if self~currentchar() <> '<' then do
- /* we found some CDATA */
- chunk~text = ''
+ /* we found some TEXT */
curline = lineidx
+ buffer1 = .MutableBuffer~new
do while eof = .false & self~currentchar() <> '<'
-- Do NOT collapse the white space and newlines out of the chunk!
-- We leave that task up to the client of this class.
- -- Instead, we return a chunk with .endofline stuck in between
+ -- Instead, we return a chunk with self~endofline stuck in between
-- at the linebreaks.
- chunk~text = chunk~text || self~getchar()
+ buffer1~append(self~getchar())
if curline <> lineidx then do
- chunk~text = chunk~text || .endofline
curline = lineidx
end
end
- chunk~text = self~xlatetext(chunk~text)
- if chunk~text~strip <> '' then ,
+ chunk~text = self~xlatetext(buffer1)~string
+ -- if chunk~text~strip <> '' then , -- never strip a TEXT
self~text(chunk) -- call the public override method
return chunk
end
/* we found an XML tag, process it */
character = self~getchar() -- skip the '<'
-element = ''
curline = lineidx
nestlevel = 0
+buffer2 = .MutableBuffer~new
+do i=0 to 7
+ buffer2~append(self~currentchar(i))
+end
+iscdata = (buffer2~string == "![CDATA[")
+buffer3 = .MutableBuffer~new
do while eof = .false
- if element~substr(1, 1) = '!' then do
+ if \iscdata & buffer3~substr(1, 1) = '!' then do
-- It is possible for tags to be contained within other tags in XML
-- processing tags. The next two IF statements take care of that nesting
-- possibility. It will be up to the user to parse out the contained
@@ -220,61 +271,87 @@
if self~currentchar() = '<' then nestlevel = nestlevel + 1
if self~currentchar() = '>' & level > 0 then nestlevel = nestlevel - 1
end
- element = element || self~getchar()
+ buffer3~append(self~getchar())
if curline <> lineidx then do
- element = element || ' '
curline = lineidx
end
- if self~currentchar() = '>' & nestlevel = 0 then leave
+ if self~currentchar() = '>' then do
+ if iscdata then do
+ if self~currentchar(-2) || self~currentchar(-1) == "]]" then leave
+ end
+ else if nestlevel = 0 then leave
+ end
+ if eof = .true then do
+ errortxt = 'Error line' errlineidx': EOF within an XML tag.'
+ self~error(self~create_error(errortxt, errlineidx, errcharidx)) -- call the public override method
+ return .nil
+ end
end
-if eof = .true then do
- errortxt = 'Error line' errlineidx': EOF within an XML tag.'
- self~error(self~create_error(errortxt, errlineidx, errcharidx)) -- call the public override method
- return .nil
- end
-element = element~strip()
+raw_element = buffer3~string
+if \iscdata then buffer3~changeStr(self~endofline, " "~copies(self~endofline~length)) -- replace endofline by space(s)
+element = buffer3~string~strip()
+if self~debug then call lineout "stderr", errlineidx":"currentSourceLine()":element=["element"]"
select
when element~substr(1, 1) = '/' then do
chunk~tag = element~substr(2)
+ chunk~text = '' -- empty ==> </tag> will be displayed
self~end_element(chunk) -- call the public override method
end
when pos('?', element~substr(1, 1)) > 0 then do
chunk~tag = ''
- chunk~text = element
+ chunk~text = raw_element
self~passthrough(chunk) -- call the public override method
end
when pos('!--', element~substr(1, 3)) > 0 then do
chunk~tag = ''
- chunk~text = element
+ chunk~text = raw_element
self~passthrough(chunk) -- call the public override method
end
when pos('!', element~substr(1, 1)) > 0 then do
chunk~tag = ''
- chunk~text = element
+ chunk~text = raw_element
self~passthrough(chunk) -- call the public override method
end
when pos(element~substr(1, 1), xrange('a', 'z') || xrange('A', 'Z')) > 0 then do
parse var element tag element
+ element = element~strip()
+ if self~debug then
+ do
+ call lineout "stderr", errlineidx":"currentSourceLine()":tag =["tag"]"
+ call lineout "stderr", errlineidx":"currentSourceLine()":element=["element"]"
+ end
-- the following is needed in case the '/' immediately follows the tag
if tag~pos('/') > 0 then do
tag = tag~substr(1, tag~length - 1)
element = '/' element
end
chunk~tag = tag
+ chunk~text = raw_element
/* process the attributes */
if element~length > 0 then chunk~attr = .directory~new
do while element~length() > 0
+ -- To increase the chances of parsing success, remove all the spaces before the first quote, if any
+ -- before: a = " value of a " b = ' value of b '
+ -- after : a=" value of a " b = ' value of b '
+ firstQuote = element~pos('"')
+ firstApos = element~pos("'")
+ if firstQuote <> 0 & firstApos <> 0 then pos = min(firstQuote, firstApos)
+ else if firstQuote <> 0 then pos = firstQuote
+ else if firstApos <> 0 then pos = firstApos
+ else pos = 0
+ if pos <> 0 then element = element~left(pos - 1)~space(0) || element~substr(pos)
+ if self~debug then call lineout "stderr", errlineidx":"currentSourceLine()":element=["element"]"
if pos('="', element~word(1)) > 0 then do
parse var element attrname '="' attrvalue '"' element
attrname = attrname~strip()
- attrvalue = attrvalue~strip()
+ --attrvalue = attrvalue~strip() -- the value between quotes must remain as-is
attrvalue = self~xlatetext(attrvalue)
chunk~attr[attrname] = attrvalue
end
else if pos("='", element~word(1)) > 0 then do
parse var element attrname "='" attrvalue "'" element
attrname = attrname~strip()
- attrvalue = attrvalue~strip()
+ --attrvalue = attrvalue~strip() -- the value between quotes must remain as-is
attrvalue = self~xlatetext(attrvalue)
chunk~attr[attrname] = attrvalue
end
@@ -294,6 +371,9 @@
self~start_element(chunk) -- call the public override method
if attrname = '/' then do
endchunk = .xmlchunk~new
+ endchunk~line = lineidx
+ endchunk~col = charidx
+ endchunk~text = raw_element -- won't be printed as-is (because already printed by start_element), but will be used to decide to print nothing instead of </tag>
endchunk~tag = tag
self~end_element(endchunk) -- call the public override method
end
@@ -309,6 +389,8 @@
character = self~getchar() -- skip the '>'
return chunk
+currentSourceLine:
+return SIGL
/*----------------------------------------------------------------------------*/
/*----------------------------------------------------------------------------*/
@@ -400,11 +482,13 @@
use strict arg src
eof = .false
/* make sure this is an xml document */
+/* Not applicable to the SGML files of oorexx
if src[1]~pos('<?xml') <> 1 then do
errortxt = 'Error: Invalid XML document.'
self~error(self~create_error(errortxt, 1, 1))
- return
+ return ""
end
+*/
/* parse the xml array */
lineidx = 1
charidx = 1
@@ -430,6 +514,7 @@
return errortxt
end
lines = tfile~arrayin()
+self~detect_eol_style(tfile) -- don't assume that the file has the eol style of the current platform
tfile~close()
errortxt = self~parse_array(lines)
@@ -462,8 +547,9 @@
::method text attribute -- the text
::method tag attribute -- the xml tag name
::method attr attribute -- the tag attributes
+::attribute line -- the start line of the chunk
+::attribute col -- the start col of the chunk
-
/*----------------------------------------------------------------------------*/
/* Method: init */
/* Description: instance initialization */
@@ -481,6 +567,8 @@
self~attr = .nil -- For the start_element method this is an ooRexx directory
-- class instance. Each attribute and value is contained in
-- the ooRexx directory instance.
+self~line = 0
+self~col = 0
return
Modified: incubator/DocMusings/railroad/_diary.txt
===================================================================
--- incubator/DocMusings/railroad/_diary.txt 2009-07-13 00:37:08 UTC (rev 4911)
+++ incubator/DocMusings/railroad/_diary.txt 2009-07-13 01:47:23 UTC (rev 4912)
@@ -1,4 +1,37 @@
===============================================================================
+2009 july 11
+
+I have a small documentation which gives examples of railroads generated
+from DITA syntax diagrams, but I can't remember from where I downloaded it...
+In the meantime, remember that :
+http://www.ditainfocenter.com/eclipsehelp/index.jsp?topic=/org.ditausers.infomanager.LangSpec1.1/common/pr-d.html
+http://en.wikipedia.org/wiki/Syntax_diagram
+http://www.cs.man.ac.uk/~pjj/bnf/bnf.html
+http://ditaa.sourceforge.net/
+
+
+===============================================================================
+2009 june 29
+
+generate.rex :
+No need to iterate over the svg files. Batik rasterizer does that automatically.
+
+I find that the letters of the keywords are a little bit blurred.
+--> Changed the CSS stylesheet to declare stroke-width: 0.0; instead of 0.1
+Since the look is better, changed all the 0.1 to 0.0
+
+Another possible improvement : in constants.xml, there is a suggestion to try
+diagram_scale=0.28
+diagram_unit=in
+This is lot more CPU intensive, and I needed to declare a bigger heap for java.
+The resolution is better but the png are bigger and very slow to display.
+--> not activated for now.
+
+To investigate : I see that integer.png has small boxes floating near the symbols.
+It's probably because of <synnote>.
+
+
+===============================================================================
2009 june 28
About Clapham :
Modified: incubator/DocMusings/railroad/generate.rex
===================================================================
--- incubator/DocMusings/railroad/generate.rex 2009-07-13 00:37:08 UTC (rev 4911)
+++ incubator/DocMusings/railroad/generate.rex 2009-07-13 01:47:23 UTC (rev 4912)
@@ -1,3 +1,5 @@
+if arg(1) = "svg" then signal svg
+
-- Generate railroads using Clapham
'java net.hydromatic.clapham.Clapham -d clapham_bnf declarations.bnf'
'java net.hydromatic.clapham.Clapham -d clapham_wsn declarations.wsn'
@@ -2,2 +4,3 @@
+svg:
-- Generate railroads using com.moldflow.dita.syntaxdiagram2svg
@@ -12,10 +15,6 @@
'declarations.xml',
'>'outputdir'/debug.txt 2>&1'
-call SysFileTree outputdir"/*.svg", "svgfiles", "FO"
BATIK_ROOT = value("BATIK_ROOT",,"ENVIRONMENT")
-do i=1 to svgfiles.0
- -- of course, launching the Java VM for each image is not the most efficient technique...
- 'java -jar "'BATIK_ROOT'/extensions/batik-rasterizer-ext.jar" -onload "'svgfiles.i'"'
-end
+'java -Xmx1024M -jar "'BATIK_ROOT'/extensions/batik-rasterizer-ext.jar" -onload 'outputdir'/*.svg'
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg/css/syntaxdiagram.css
===================================================================
--- incubator/DocMusings/railroad/syntaxdiagram2svg/css/syntaxdiagram.css 2009-07-13 00:37:08 UTC (rev 4911)
+++ incubator/DocMusings/railroad/syntaxdiagram2svg/css/syntaxdiagram.css 2009-07-13 01:47:23 UTC (rev 4912)
@@ -28,7 +28,7 @@
font-weight: bold;
font-size: 8pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] g[class~="var"] text {
@@ -38,7 +38,7 @@
font-style: italic;
font-size: 8pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] g[class~="delim"] text {
@@ -49,7 +49,7 @@
font-weight: normal;
font-size: 8pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] g[class~="oper"] text {
@@ -60,7 +60,7 @@
font-weight: normal;
font-size: 8pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] g[class~="sep"] text {
@@ -71,7 +71,7 @@
font-weight: normal;
font-size: 8pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] g[class~="repsep"] text {
@@ -82,7 +82,7 @@
font-weight: normal;
font-size: 8pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] a[class~="fragref"] text {
@@ -93,7 +93,7 @@
font-weight: normal;
font-size: 8pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] g[class~="note"] text {
@@ -102,7 +102,7 @@
font-weight: normal;
font-size: 6pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] g[class~="boxed"] > g[class~="note"] text {
@@ -111,7 +111,7 @@
font-weight: normal;
font-size: 6pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] a[class~="note"] text {
@@ -122,7 +122,7 @@
font-weight: normal;
font-size: 6pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
svg[class="syntaxdiagram"] a[class~="boxed"] > g[class~="note"] text {
@@ -133,5 +133,5 @@
font-weight: normal;
font-size: 6pt;
stroke: #000000;
- stroke-width: 0.1;
+ stroke-width: 0.0;
}
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/DO-conditional.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/DO-repetitor.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/DO.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/Decimal integer.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/Hexadecimal literal.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/IPV4 address in dot form-IPv4-range integer.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/IPV4 address in dot form-decimal digit.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/IPV4 address in dot form.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/MathLoadFuncs.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/RxCalcSqrt.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/URL syntax 1.png
===================================================================
(Binary files differ)
Modified: incubator/DocMusings/railroad/syntaxdiagram2svg_out/URL syntax 2.png
===================================================================
(Binary files differ)
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|