This is eText.TaggedText.m in view mode; [Download] [Up]
{\rtf0\ansi{\fonttbl\f0\fmodern Courier;\f1\ftech Symbol;\f2\fmodern Ohlfs;}
\paperw11640
\paperh8400
\margl40
\margr40
{\colortbl;\red0\green0\blue0;}
\pard\tx520\tx1060\tx1600\tx2120\tx2660\tx3200\tx3720\tx4260\tx4800\tx5320\f0\b0\i0\ulnone\fs24\fc0\cf0 //ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ\
//
\i
\b FILENAME
\b0 :
\b\i0 eText.TaggedText.m\
\b0 //
\i
\b SUMMARY
\b0 :
\b\i0 Implementation of Tagged markup formats of eText (HTML, LaTeX)
\b0 \
//
\b\i CATEGORY
\b0 :
\i0
\b TaggedText
\b0 \
//
\b\i PROTOCOLS
\b0 :
\i0
\b NXRegisterPrintfProc()
\b0 \
//
\b\i INTERFACE
\b0 :
\i0
\b See ChooseEncoding.Tool
\b0 \
//
\b\i AUTHOR
\b0 :
\b\i0 Rohit Khare and Tom Zavisca
\b0 \
//
\b\i COPYRIGHT
\b0 :
\f1\i0 Ó
\f0\b 1993,94 California Institure of Technology, eText Project\
\b0 //ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ\
//
\b\i Implementation Comments\
\b0\i0 // There's a lot of malarkey involved with the API for the encoders and\
// the printfProc registrations. All the "user" needs is
\b + flushHTMLEncoding
\b0 .\
//\
//
\b currentHTMLEncoding
\b0 is a file-global
\b char**
\b0 of
\b ENTITIES
\b0 entries\
//
\b currentHTMLEncodingLength
\b0 is a file-global unsigned
\b char
\b0 [
\b ENTITIES
\b0 ];\
//
\b defaultHTMLEncoding
\b0 is a
\b char*
\b0 [
\b ENTITIES
\b0 ] C array.\
// An attempt is made to read encodings from the file specified in\
// a user dwrite.\
//\
// Something that bothers me about encoders: how can we properly use\
// Symbol font?
\f1 S ¹
\f0 \\sigma ?\
//\
// After investigating the interactions of Annotations and Tagging, \
// I have decided that it is not neccessary to return to ground state\
// before writing HTML for an annotation. The physical and logical tags\
// apply to the run containing the annotation as well.\
// Reason: link button on same line as an H3 font descrip.\
// Output: close, linkbutton, open, desc, close splits into two lines.\
//ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ\
//
\b\i History
\b0\i0 \
// 11/04/94:
\b DESIGN REV: HTML annotations don't start from ground state
\b0 \
// 10/17/94:
\b Cleaned up for eText5.
\b0 \
// 08/05/94:
\b Completely Rearchitected for 5.0. RK
\b0 \
//ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ\
//
\b Imported Interfaces
\b0 \
//\
#import "
\b eText.TaggedText.h
\b0 "\
#import <
\b ctype.h
\b0 >\
\
\i @implementation eText(TaggedText)\
\i0 //ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ\
//
\b Stream Operators
\b0 \
//\
-
\b writeHTML
\b0 :(
\b NXStream
\b0 *) s
\b withTags
\b0 :(
\b taggingInfo
\b0 *) tags \{\
\b int
\b0 k,N;\
\b NXRun
\b0 *
\b curr
\b0 ;\
\b NXTextBlock
\b0 *
\b currBlock
\b0 ;\
int
\b currentOffset
\b0 ,
\b targetOffset
\b0 ;\
\b NXAtom
\b0
\b closer
\b0 ;\
\b taggingInfo
\b0 *
\b aTag
\b0 ,*
\b found
\b0 ;\
\b taggingInfo
\b0
\b fakeTag
\b0 ;\
\b id
\b0
\b fm
\b0 ; //
\i FontManager
\i0 \
\
\b N
\b0 =
\b theRuns
\b0 ->chunk.
\b used
\b0 /sizeof(NXRun);\
\b curr
\b0 = theRuns->
\b runs
\b0 ;\
\b currBlock
\b0 = [self
\b firstTextBlock
\b0 ];\
\b currentOffset
\b0 =
\b 0
\b0 ;\
\b closer
\b0 =
\b NULL
\b0 ;\
\b fm
\b0 = [
\b FontManager
\b0 new];\
\
for (k=0;
\b k < N
\b0 ; k++) \{\
if (curr->
\b info
\b0 ==
\b NULL
\b0 ) \{\
\i // Encode the state for this run.\
// First, is this a tagged run? If so, do we need to close the\
// previous state?\
\i0 \
\b aTag
\b0 =
\b tags
\b0 ;
\b found
\b0 =
\b NULL
\b0 ;\
while (
\b !found
\b0 && aTag &&
\b aTag->font
\b0 ) \{\
if (
\b curr->font
\b0 ==
\b aTag->font
\b0 )
\b found
\b0 = aTag; \
aTag++;\
\}\
\
if (
\b !found
\b0 ) \{ //
\i search for physical tags\
\i0
\b char
\b0 tmp[32],*
\b family
\b0 ;\
\b NXFontTraitMask
\b0
\b traits
\b0 ;\
int
\b weight
\b0 ;\
float
\b size
\b0 ;\
BOOL
\b isFixedPitch
\b0 ;\
\
*tmp = 0;\
\b isFixedPitch
\b0 = //
\i primitive monospacing test\
\i0 ([curr->font
\b metrics
\b0 ])
\fc1\cf1 ->
\b isFixedPitch
\b0\fc0\cf0 ;\
[fm
\b getFamily
\b0 :&family
\b traits
\b0 :&traits
\b weight
\b0 :&weight \
\b size
\b0 :&size
\b ofFont
\b0 :curr->font];\
if (
\b isFixedPitch
\b0 ) strcat(tmp, "
\b <TT>
\b0 ");\
if (curr->rFlags.
\b underline
\b0 ) strcat(tmp, "
\b <U>
\b0 ");\
if (traits &
\b NX_BOLD
\b0 ) strcat(tmp, "
\b <B>
\b0 ");\
if (traits &
\b NX_ITALIC
\b0 ) strcat(tmp, "
\b <I>
\b0 "); \
if (*tmp) \{ //
\i don't bother unless we got styles\
\i0
\b fakeTag
\b0 .
\b start
\b0 = NXUniqueString(
\b tmp
\b0 );\
*tmp = 0;\
if (traits &
\b NX_ITALIC
\b0 ) strcat(tmp, "
\b </I>
\b0 ");\
if (traits &
\b NX_BOLD
\b0 ) strcat(tmp, "
\b </B>
\b0 ");\
if (curr->rFlags.
\b underline
\b0 ) strcat(tmp, "
\b </U>
\b0 ");\
if (
\b isFixedPitch
\b0 ) strcat(tmp, "
\b </TT>
\b0 ");\
\b fakeTag
\b0 .
\b end
\b0 = NXUniqueString(
\b tmp
\b0 );\
\b found
\b0 = &
\b fakeTag
\b0 ;\
\}\
\}\
\
if (
\b found
\b0 ) \{ //
\i stop previous tag, if differs\
\i0 if (
\b found->end != closer
\b0 ) \{\
if (closer) \{ //
\i transition
\i0 \
\b NXWrite
\b0 (s,
\b closer
\b0 , strlen(closer));\
\}\
\b NXWrite
\b0 (s, found->
\b start
\b0 , strlen(found->start));\
\b closer
\b0 = found->
\b end
\b0 ;\
\}\
\} else if (closer) \{ //
\i return to ground state\
\i0
\b NXWrite
\b0 (s,
\b closer
\b0 , strlen(closer));\
\b closer
\b0 =
\b NULL
\b0 ;\
\}\
\} else \{\
//
\i clear tagging state, write out the annotation
\i0 \
//
\i MAJOR DESIGN CHANGE: SEE HISTORY & NOTES!!!! RK, 11/4
\i0 \
// if (closer) \{\
//
\b NXWrite
\b0 (s,
\b closer
\b0 , strlen(closer));\
//
\b closer
\b0 =
\b NULL
\b0 ;\
//\}\
if ([curr->
\b info
\b0
\b respondsTo
\b0 :@selector(
\b writeHTML:forView:
\b0 )]) \{\
[curr->
\b info
\b0
\b writeHTML
\b0 :s
\b forView
\b0 :self];\
\}\
\}\
\
\i // encode the text corresponding to the run\
// misson is to write (cumulative) curr->chars chars beginning\
// at currentCount. boundaries may map onto > 1 block
\i0 \
\b targetOffset
\b0 =
\b currentOffset
\b0
\b +
\b0 curr->
\b chars
\b0 ;\
//
\i consume full blocks
\i0 \
while ((
\b currBlock
\b0 ) && (
\b targetOffset
\b0 >= (currBlock->
\b chars
\b0 ))) \{\
if (
\b !
\b0 (curr->
\b info
\b0 )) //
\i throw annotated bits in bucket
\i0 \
if(targetOffset > currentOffset) //
\i don't pass len=0 to encoder
\i0 \
\b HTMLEncoder
\b0 (s, currBlock->text+currentOffset,\
currBlock->chars - currentOffset);\
targetOffset-=currBlock->chars;\
currBlock=currBlock->next;\
currentOffset=0;\
\}\
//
\i consume partial block
\i0 \
if (
\b currBlock
\b0 && (
\b !
\b0 curr->
\b info
\b0 )) //
\i throw annotated bits in bucket\
\i0 if(targetOffset > currentOffset) //
\i don't pass len=0 to encoder
\i0 \
\b HTMLEncoder
\b0 (s, currBlock->text + currentOffset,\
targetOffset-currentOffset);\
currentOffset=targetOffset;\
curr++;\
\} \
if (closer) \{\
\b NXWrite
\b0 (s,
\b closer
\b0 , strlen(closer));\
\b closer
\b0 =
\b NULL
\b0 ;\
\}\
return self;\
\}\
\
-
\b writeLaTeX
\b0 :(
\b NXStream
\b0 *) s
\b withTags
\b0 :(
\b taggingInfo
\b0 *) tags \{\
\b int
\b0 k,N;\
\b NXRun
\b0 *
\b curr
\b0 ;\
\b NXTextBlock
\b0 *
\b currBlock
\b0 ;\
int
\b currentOffset
\b0 ,
\b targetOffset
\b0 ;\
\b NXAtom
\b0
\b closer
\b0 ,
\b oldStart
\b0 =
\b NULL
\b0 ;\
\b taggingInfo
\b0 *
\b aTag
\b0 ,*
\b found
\b0 ;\
\b taggingInfo
\b0
\b fakeTag
\b0 ;\
\b id
\b0
\b fm
\b0 ; //
\i FontManager
\i0 \
\
\b N
\b0 =
\b theRuns
\b0 ->chunk.
\b used
\b0 /sizeof(NXRun);\
\b curr
\b0 = theRuns->
\b runs
\b0 ;\
\b currBlock
\b0 = [self
\b firstTextBlock
\b0 ];\
\b currentOffset
\b0 =
\b 0
\b0 ;\
\b closer
\b0 =
\b NULL
\b0 ;\
\b fm
\b0 = [
\b FontManager
\b0 new];\
\
for (k=0;
\b k < N
\b0 ; k++) \{\
if (curr->
\b info
\b0 ==
\b NULL
\b0 ) \{\
\i // Encode the state for this run.\
// First, is this a tagged run? If so, do we need to close the\
// previous state?\
\i0 \
\b aTag
\b0 =
\b tags
\b0 ;
\b found
\b0 =
\b NULL
\b0 ;\
while (
\b !found
\b0 && aTag &&
\b aTag->font
\b0 ) \{\
if (
\b curr->font
\b0 ==
\b aTag->font
\b0 )
\b found
\b0 = aTag; \
aTag++;\
\}\
\
if (
\b !found
\b0 ) \{ //
\i search for physical tags\
\i0
\b char
\b0 tmp[32],*
\b family
\b0 ;\
\b NXFontTraitMask
\b0
\b traits
\b0 ;\
int
\b weight
\b0 ;\
float
\b size
\b0 ;\
BOOL
\b isFixedPitch
\b0 ;\
\
*tmp = 0;\
\b isFixedPitch
\b0 = //
\i primitive monospacing test\
\i0 ([curr->font
\b metrics
\b0 ])
\fc1\cf1 ->
\b isFixedPitch
\b0\fc0\cf0 ;\
[fm
\b getFamily
\b0 :&family
\b traits
\b0 :&traits
\b weight
\b0 :&weight \
\b size
\b0 :&size
\b ofFont
\b0 :curr->font];\
if (
\b isFixedPitch
\b0 ) strcat(tmp, "
\b \{\\\\tt
\b0 ");\
if (curr->rFlags.
\b underline
\b0 ) strcat(tmp, "
\b \\\\underline\{
\b0 ");\
if (traits &
\b NX_BOLD
\b0 ) strcat(tmp, "
\b \{\\\\bf
\b0 ");\
if (traits &
\b NX_ITALIC
\b0 ) strcat(tmp, "
\b \{\\\\it
\b0 "); \
if (*tmp) \{ //
\i don't bother unless we got styles\
\i0
\b fakeTag
\b0 .
\b start
\b0 = NXUniqueString(
\b tmp
\b0 );\
*tmp = 0;\
if (traits &
\b NX_ITALIC
\b0 ) strcat(tmp, "
\b \}
\b0 ");\
if (traits &
\b NX_BOLD
\b0 ) strcat(tmp, "
\b \}
\b0 ");\
if (curr->rFlags.
\b underline
\b0 ) strcat(tmp, "
\b \}
\b0 ");\
if (
\b isFixedPitch
\b0 ) strcat(tmp, "
\b \}
\b0 ");\
\b fakeTag
\b0 .
\b end
\b0 = NXUniqueString(
\b tmp
\b0 );\
\b found
\b0 = &
\b fakeTag
\b0 ;\
\}\
\}\
\
if (
\b found
\b0 ) \{ //
\i stop previous tag, if differs\
// how can we tell if the state has changed?\
// the assumption is that a run necessarily corresponds\
// to a change of style -- but a colorchange wouldn't.\
// with HTML, the exact closer string would be unique\
// for LaTeX the heuristic is that every _opener_ is unique\
// thus, if the opener is unchanged, we short-circuit the\
// close-reopen pair.\
\i0 if (found->
\b start
\b0
\b !=
\b0
\b oldStart
\b0 ) \{\
if (closer) \{ //
\i transition
\i0 \
\b NXWrite
\b0 (s,
\b closer
\b0 , strlen(closer));\
\}\
\b NXWrite
\b0 (s, found->
\b start
\b0 , strlen(found->start));\
\b closer
\b0 = found->
\b end
\b0 ;\
\b oldStart
\b0 = found->
\b start
\b0 ;\
\}\
\} else if (closer) \{ //
\i return to ground state\
\i0
\b NXWrite
\b0 (s,
\b closer
\b0 , strlen(closer));\
\b closer
\b0 =
\b NULL
\b0 ;\
\}\
\} else \{\
//
\i clear tagging state, write out the annotation
\i0 \
if (closer) \{\
\b NXWrite
\b0 (s,
\b closer
\b0 , strlen(closer));\
\b closer
\b0 =
\b NULL
\b0 ;\
\}\
if ([curr->
\b info
\b0
\b respondsTo
\b0 :@selector(
\b writeLaTeX:forView:
\b0 )]) \{\
[curr->
\b info
\b0
\b writeLaTeX
\b0 :s
\b forView
\b0 :self];\
\}\
\}\
\
\i // encode the text corresponding to the run\
// misson is to write (cumulative) curr->chars chars beginning\
// at currentCount. boundaries may map onto > 1 block
\i0 \
\b targetOffset
\b0 =
\b currentOffset
\b0
\b +
\b0 curr->
\b chars
\b0 ;\
//
\i consume full blocks
\i0 \
while ((
\b currBlock
\b0 ) && (
\b targetOffset
\b0 >= (currBlock->
\b chars
\b0 ))) \{\
if (
\b !
\b0 (curr->
\b info
\b0 )) //
\i throw annotated bits in bucket
\i0 \
if(targetOffset > currentOffset) //
\i don't pass len=0 to encoder
\i0 \
\b LaTeXEncoder
\b0 (s, currBlock->text+currentOffset,\
currBlock->chars - currentOffset);\
targetOffset-=currBlock->chars;\
currBlock=currBlock->next;\
currentOffset=0;\
\}\
//
\i consume partial block
\i0 \
if (
\b currBlock
\b0 && (
\b !
\b0 curr->
\b info
\b0 )) //
\i throw annotated bits in bucket\
\i0 if(targetOffset > currentOffset) //
\i don't pass len=0 to encoder
\i0 \
\b LaTeXEncoder
\b0 (s, currBlock->text + currentOffset,\
targetOffset-currentOffset);\
currentOffset=targetOffset;\
curr++;\
\} \
if (closer) \{\
\b NXWrite
\b0 (s,
\b closer
\b0 , strlen(closer));\
\b closer
\b0 =
\b NULL
\b0 ;\
\}\
return self;\
\}\
\
//ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ\
//
\b String Operators
\b0 \
//\
+
\b encodeHTML
\b0 :(NXStream *) s
\b from
\b0 : (unsigned char *) theChars
\b length
\b0 : (int)len \{\
\b HTMLEncoder
\b0 (s,theChars,len); return self;\}\
+
\b encodeLaTeX
\b0 :(NXStream *) s
\b from
\b0 : (unsigned char *) theChars
\b length
\b0 :(int)len \{ \
\b LaTeXEncoder
\b0 (s,theChars,len); return self;\}\
+
\b encodeURI
\b0 :(NXStream *) s
\b from
\b0 : (unsigned char *) theChars
\b length
\b0 :(int)len \{ \
\b URIEncoder
\b0 (s,theChars,len); return self;\}\
\
//ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ\
//
\b HTMLEncoder API
\b0 \
//\
+
\b flushHTMLEncoding
\b0 \{\
//
\i Next access will force reloading according to UserModel
\i0 \
\b currentHTMLEncoding
\b0 =
\b NULL
\b0 ; return self;\
\}\
\
@end\
\
//ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ\
//
\b Encoder API
\b0 \
//
\i note that these are file-globals, and thus apply to the
\i0 \
//
\i entire eText process; encodings are not chosen on a per-document basis
\i0 .\
//\
\
const char *
\b defaultHTMLEncoding
\b0 [
\b ENTITIES
\b0 ];\
char **
\b currentHTMLEncoding
\b0 =NULL;\
char *
\b currentBuffer
\b0 ;\
unsigned char *
\b currentHTMLEncodingLength
\b0 ;\
\
void
\b HTMLEncoder
\b0 (NXStream *
\b stream
\b0 , unsigned char *
\b item
\b0 , int
\b len
\b0 ) \{\
int
\b i
\b0 ;\
\
if(
\b !currentHTMLEncoding
\b0 ) \{ //
\i Hence the
\b +flushHTMLEncoding\
\b0\i0 char
\b HTMLResourceFilePath
\b0 [MAXPATHLEN];\
\
[[NXBundle
\b mainBundle
\b0 ]
\b getPath
\b0 :
\b HTMLResourceFilePath
\b0 \
\b forResource
\b0 :[userModel stringQuery:"
\b HTMLEncoding
\b0 "]\
\b ofType
\b0 :
\b ENCD_EXT
\b0 ];\
if(*HTMLResourceFilePath) \{ //
\i try to load from this path\
\i0 unsigned char *
\b tempBuffer
\b0 ;\
unsigned char *
\b tempLengths
\b0 ;\
char **
\b tempEncoding
\b0 ;\
\
\b tempBuffer
\b0 =
\b malloc
\b0 (
\b 4*MAXPATHLEN
\b0 *sizeof(unsigned char));\
//
\i The above is C Programmer's Disease
\i0 \
\b tempLengths
\b0 = malloc(
\b ENTITIES
\b0 * sizeof(
\b unsigned char
\b0 ));\
\b tempEncoding
\b0 = malloc(
\b ENTITIES
\b0 * sizeof(
\b char *
\b0 ));\
if (
\b readEncodingTableFromFile
\b0 (\
\b HTMLResourceFilePath
\b0 ,
\b tempEncoding
\b0 ,
\b tempLengths
\b0 ,
\b tempBuffer
\b0 ))\{\
//
\i no freeing if not defaultEncoding
\i0 \
//
\i is a memory leak noone cares about
\i0 \
\b currentHTMLEncodingLength
\b0 =
\b tempLengths
\b0 ;\
\b currentHTMLEncoding
\b0 =
\b tempEncoding
\b0 ;\
\b currentBuffer
\b0 =
\b tempBuffer
\b0 ;\
\} else \{\
\b free
\b0 (tempBuffer);
\b free
\b0 (tempLengths);
\b free
\b0 (tempEncoding);\
tempBuffer = tempLengths = tempEncoding = NULL;\
NXLogError("
\b Could not read encoding data from %s
\b0 ",\
\b HTMLResourceFilePath
\b0 );\
\}\
\} \
\}\
\
if (
\b !currentHTMLEncoding
\b0 ) \{ //
\i Error fall-through\
\i0 //
\i "use" the defaultEncoding
\i0 .\
currentHTMLEncoding =
\b defaultHTMLEncoding
\b0 ;\
currentHTMLEncodingLength=malloc(ENTITIES * sizeof(unsigned char));\
for(i=0;i<ENTITIES;i++)\
\b currentHTMLEncodingLength
\b0 [i] = \
(
\b currentHTMLEncoding
\b0 [i] ?
\b strlen
\b0 (
\b currentHTMLEncoding
\b0 [i]) :
\b 0
\b0 );\
\}\
\
//
\i The two "modes" of the Encoder, using the userData parameter
\i0 \
if(
\b !len
\b0 ) len=
\b strlen
\b0 ((unsigned char*)
\b item
\b0 );\
\
for (i=0; i<len; i++) \{\
\b NXWrite
\b0 (
\b stream
\b0 ,
\b currentHTMLEncoding
\b0 [(unsigned char)
\b item
\b0 [i]],\
\b currentHTMLEncodingLength
\b0 [(unsigned char)
\b item
\b0 [i]]);\
\}\
\}\
\
void
\b URIEncoder
\b0 (NXStream *
\b s
\b0 , unsigned char *
\b item
\b0 , int
\b len
\b0 ) \{\
int i;\
\
//
\i The two "modes" of the Encoder, using the userData parameter
\i0 \
if(
\b !len
\b0 ) len=
\b strlen
\b0 ((unsigned char*)
\b item
\b0 );\
\
for (i=0; i<len; i++) \{\
unsigned ch = item[i];\
if (
\b isalnum
\b0 (ch) || ((ch=='
\b %
\b0 ')&&
\b isdigit
\b0 (item[
\b i+1
\b0 ])&&
\b isdigit
\b0 (item[
\b i+2
\b0 ])))\
\b NXPutc
\b0 (s,
\b ch
\b0 );\
else
\b switch
\b0 (
\b ch
\b0 ) \{\
case '
\b :
\b0 ': \
case '
\b /'
\b0 : \
case '
\b \\\\
\b0 ': \
//
\i "safe" in RFC1630 BNF\
\i0 case '
\b $
\b0 ': \
case '
\b -
\b0 ': \
case '
\b _
\b0 ': \
case '
\b @
\b0 ': \
case '
\b .
\b0 ': \
case '
\b &
\b0 ': \
case '+':\
//
\i "extra" in RFC1630 BNF\
\i0 case '
\b !
\b0 ': \
case '
\b *
\b0 ': \
case '\\
\b "
\b0 ': \
case '\\
\b '
\b0 ': \
case '
\b |
\b0 ': \
case '
\b ,
\b0 ':\
NXPutc(s,
\b ch
\b0 ); break;\
\b default
\b0 : //
\i encode as %hex\
\i0 NXPrintf(s,"%%%x",
\b ch
\b0 ); break;\
\}\
\}\
\}\
\
void
\b LaTeXEncoder
\b0 (NXStream *
\b s
\b0 , unsigned char *
\b item
\b0 , int
\b len
\b0 ) \{\
int i;\
\
//
\i The two "modes" of the Encoder, using the userData parameter
\i0 \
if(
\b !len
\b0 ) len=
\b strlen
\b0 ((unsigned char*)
\b item
\b0 );\
\
for (i=0; i<len; i++) \{\
unsigned ch = item[i];\
\b switch
\b0 (
\b ch
\b0 ) \{\
case '
\b <
\b0 ': NXWrite(s, "
\b $<$
\b0 ",3); break;\
case '
\b >
\b0 ': NXWrite(s, "
\b $>$
\b0 ",3); break;\
case '
\b \\\\
\b0 ': NXWrite(s, "
\b $\\\\backslash$
\b0 ",12); break;\
case '
\b ~
\b0 ': NXWrite(s, "
\b \\\\~
\b0 ",2); break;\
case '
\b ^
\b0 ': NXWrite(s, "
\b \\\\^
\b0 ",2); break;\
case '
\b \{
\b0 ': NXWrite(s, "
\b \\\\\{
\b0 ",2); break;\
case '
\b \}
\b0 ': NXWrite(s, "
\b \\\\\}
\b0 ",2); break;\
case '
\b %
\b0 ': NXWrite(s, "
\b \\\\%
\b0 ",2); break;\
case '
\b #
\b0 ': NXWrite(s, "
\b \\\\#
\b0 ",2); break;\
case '
\b _
\b0 ': NXWrite(s, "
\b \\\\_
\b0 ",2); break;\
case '
\b &
\b0 ': NXWrite(s, "
\b \\\\&
\b0 ",2); break;\
case '
\b $
\b0 ': NXWrite(s, "
\b \\\\$
\b0 ",2); break;\
case '
\b \\n
\b0 ': NXWrite(s, "
\b \\\\par\\n
\b0 ",1); break;\
case '
\b \\t
\b0 ': //
\i we should do something here for tabs
\i0 \
\b default
\b0 : //
\i we should do something here for extended symbols\
\i0 NXPutc(s,
\b ch
\b0 ); break;\
\}\
\}\
\}\
\
BOOL
\b readEncodingTableFromFile
\b0 (const char *
\b path
\b0 , char **
\b targetEncoding
\b0 ,\
unsigned char *
\b targetLengths
\b0 ,char*
\b targetBuffer
\b0 ) \{\
\b NXStream
\b0 *
\b s
\b0 ;\
int i,j,len,maxlen;\
char *theChars,*current;\
\
\b s
\b0 =
\b NXMapFile
\b0 (
\b path
\b0 ,
\b NX_READONLY
\b0 );\
if(s) \{\
\b NXGetMemoryBuffer
\b0 (
\b s
\b0 ,&theChars,&len, &maxlen);\
i=j=0;\
while (
\b i<ENTITIES
\b0 && (j < len)) \{\
while (theChars[j] == '
\b #
\b0 ') \{ //
\i consume comment lines\
\i0
\b while
\b0 ((j<len) && (
\b theChars
\b0 [
\b j++
\b0 ] != '
\b \\n
\b0 ')); \
\}\
if (theChars[j++] == '
\b \\"
\b0 ')\{ //
\i we have a winner!\
\i0 current = targetBuffer;\
\b while
\b0 (theChars[j]
\b !=
\b0 '
\b \\"
\b0 ')\{\
//
\i heuristics identical to NXStringTable
\i0 \
\b switch
\b0 (theChars[j]) \{\
case '
\b \\\\
\b0 ' : \
switch (theChars[++j]) \{\
case '
\b n
\b0 ' : *(targetBuffer++)= '
\b \\n
\b0 '; break;\
case '
\b t
\b0 ' : *(targetBuffer++)= '
\b \\t
\b0 '; break;\
case '
\b \\\\
\b0 ' : *(targetBuffer++)= '
\b \\\\
\b0 '; break;\
case '
\b \\"
\b0 ' : *(targetBuffer++)= '
\b \\"
\b0 '; break;\
case '
\b a
\b0 ' : *(targetBuffer++)= '
\b \\a
\b0 '; break;\
case '
\b b
\b0 ' : *(targetBuffer++)= '
\b \\b
\b0 '; break;\
case '
\b f
\b0 ' : *(targetBuffer++)= '
\b \\f
\b0 '; break;\
case '
\b r
\b0 ' : *(targetBuffer++)= '
\b \\r
\b0 '; break;\
case '
\b v
\b0 ' : *(targetBuffer++)= '
\b \\v
\b0 '; break;\
\b default
\b0 : *(targetBuffer++)=
\b theChars
\b0 [j]; break;\
\} break;\
\b default
\b0 : *(targetBuffer++) =
\b theChars
\b0 [j]; break;\
\}\
j++;\
\}\
*(targetBuffer++)=0;\
\b targetEncoding
\b0 [i]=
\b current
\b0 ;\
\b targetLengths
\b0 [i]=
\b strlen
\b0 (
\b targetEncoding
\b0 [i]);\
i++;\
\}\
\b while
\b0 ((j<len) && (
\b theChars
\b0 [
\b j++
\b0 ]
\b !=
\b0 '
\b \\n
\b0 ')); //
\i consume until EOL
\i0 \
\}\
\b NXCloseMemory
\b0 (s,
\b NX_FREEBUFFER
\b0 );\
return
\b YES
\b0 ;\
\}\
return
\b NO
\b0 ;\
\}\
\
\
const char *
\b defaultHTMLEncoding
\b0 [
\b ENTITIES
\b0 ] = \{\
"", /* NUL */\
"", /* SOH */\
"", /* STX */\
"", /* ETX */\
"", /* EOT */\
"", /* ENQ */\
"", /* ACK */\
"", /* BEL */\
"", /* BS */\
"\\t", /* TAB */\
"<BR>\\n", /* NEWLINE */\
"", /* VT */\
"", /* FF */\
"\\r", /* CR */\
"", /* SO */\
"", /* SI */\
"", /* DLE */\
"", /* DC1 (XON)*/\
"", /* DC2 */\
"", /* DC3 (XOFF)*/\
"", /* DC4 */\
"", /* NAK */\
"", /* SYN */\
"", /* ETB */\
"", /* CAN */\
"", /* EM */\
"", /* SUB */\
"", /* ESC */\
"", /* FS */\
"", /* GS */\
"", /* RS */\
"", /* US */\
" ", /* SPACE */\
"!",\
""",\
"#",\
"$",\
"%",\
"&",\
"'",\
"(",\
")",\
"*",\
"+",\
",",\
"-",\
".",\
"/",\
"0",\
"1",\
"2",\
"3",\
"4",\
"5",\
"6",\
"7",\
"8",\
"9",\
":",\
";",\
"<",\
"=",\
">",\
"?",\
"@",\
"A",\
"B",\
"C",\
"D",\
"E",\
"F",\
"G",\
"H",\
"I",\
"J",\
"K",\
"L",\
"M",\
"N",\
"O",\
"P",\
"Q",\
"R",\
"S",\
"T",\
"U",\
"V",\
"W",\
"X",\
"Y",\
"Z",\
"[",\
"\\\\",\
"]",\
"^",\
"_",\
"`",\
"a",\
"b",\
"c",\
"d",\
"e",\
"f",\
"g",\
"h",\
"i",\
"j",\
"k",\
"l",\
"m",\
"n",\
"o",\
"p",\
"q",\
"r",\
"s",\
"t",\
"u",\
"v",\
"w",\
"x",\
"y",\
"z",\
"\{",\
"|",\
"\}",\
"~",\
"", /* DEL */\
" ",\
"À",\
"Á",\
"Â",\
"Ã",\
"Ä",\
"Å",\
"Ç",\
"È",\
"É",\
"Ê",\
"Ë",\
"Ì",\
"Í",\
"Î",\
"Ï",\
"Ð", // 0x90\
"Ñ",\
"Ò",\
"Ó",\
"Ô",\
"Õ",\
"Ö",\
"Ù",\
"Ú",\
"Û",\
"Ü",\
"Ý",\
"Þ",\
"mu",\
" x ",\
" / ",\
"(c)", // 0xA0\
" ! ",\
" cents ",\
" Pound ",\
"/",\
" Yen ",\
" Florin ",\
" Section ",\
" Currency ",\
"'",\
"``",\
"<<",\
"<",\
">",\
"fi",\
"fl",\
"(R)", // 0xB0\
"-",\
" * ", // dagger\
" ** ", // dubdagger\
" . ", // centered period\
" | ", // broken pipe\
" P ", // Paragraph\
"*", // bullet\
",", // low-quote\
",,", // low dubquote\
"''", // up dub\
">>",\
"...",\
"%%", // per thousand\
"!", // not!\
"?", // upside down ? \
"1", // 0xC0\
"`",\
"'",\
"^",\
"~",\
"-", // macron\
"\\\\/", // breve\
".", \
"..", // uml\
"2",\
"o",\
",", // cedilla\
"3",\
"''", \
",", // backward cedilla\
"\\\\/", // caron\
"--", // 0xD0\
"+/-",\
"(1/4)",\
"(1/2)",\
"(3/4)",\
"à",\
"á",\
"â",\
"ã",\
"ä",\
"å",\
"ç",\
"è",\
"é",\
"ê",\
"ë",\
"ì", //0xE0\
"Æ",\
"í",\
"a",\
"î",\
"ï",\
"ð",\
"ñ",\
"L",\
"Ø",\
"OE",\
"o",\
"ò",\
"ó", \
"ô",\
"õ",\
"ö",\
"æ", \
"ù",\
"ú",\
"û", \
"i",\
"ü",\
"ý",\
"l", \
"ø",\
"oe",\
"B",\
"þ", \
"ÿ",\
"",\
""\
\};
}
These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Netfuture.ch.