r5238 MediaWiki - Code Review archive

Repository:	MediaWiki
Revision:	< r5237‎ \| r5238 \| r5239 >
Date:	21:11, 15 September 2004
Author:	timwi
Status:	old
Tags:
Comment:	several bugfixes, esp. with apostrophes (bold/italics) and headings. The ENDHEADING token now occurs only in the "right" places, so it needn't be text any longer.
Modified paths:	/trunk/flexbisonparse/parsetree.c (modified) (history) /trunk/flexbisonparse/parsetree.h (modified) (history) /trunk/flexbisonparse/wikilex.l (modified) (history) /trunk/flexbisonparse/wikiparse.y (modified) (history)

Diff [purge]

Index: trunk/flexbisonparse/parsetree.h
—	—	@@ -115,6 +115,7 @@
116	116	Node convertTableRowToText (int info);
117	117	Node convertTableCellToText (int info);
118	118	Node convertTableHeadToText (int info);
	119	+Node convertHeadingToText (int info);
119	120
120	121	/* Parameter must be a TextBlock. Turns something like
121	122	* <italics>X<italics>Y</italics>Z</italics> into
Index: trunk/flexbisonparse/wikiparse.y
—	—	@@ -89,6 +89,13 @@
90	90	\| table zeroormorenewlines { debugf ("blocknotbl#5 "); $$ = $1; }
91	91	\| comment zeroormorenewlines { debugf ("blocknotbl#6 "); $$ = $1; }
92	92
	93	+heading : HEADING text ENDHEADING
	94	+ { debugf ("heading#1 "); $$ = nodeAddChild (newNodeI (Heading, $1), $2); }
	95	+ \| HEADING text /* for eof */
	96	+ { debugf ("heading#2 "); $$ = nodeAddChild (newNode (Paragraph), makeTextBlock (convertHeadingToText ($1), $2)); }
	97	+ \| HEADING
	98	+ { debugf ("heading#3 "); $$ = nodeAddChild (newNode (Paragraph), convertHeadingToText ($1)); }
	99	+
93	100	preblock : preline { debugf ("preblock#1 "); $$ = nodeAddChild (newNode (PreBlock), $1); }
94	101	\| preblock preline { debugf ("preblock#2 "); $$ = nodeAddChild ($1, $2); }
95	102
—	—	@@ -225,20 +232,29 @@
226	233	(TextToken, "'"), nodeAddChild (newNode (Italics), $2))); }
227	234	\| APO3 textnobold
228	235	{ debugf ("italicsorbold#9 "); $$ = makeTextBlock (newNodeS (TextToken, "'''"), $2); }
	236	+ \| APO5 textnoboit APO5
	237	+ { debugf ("italicsorbold#10 "); $$ = nodeAddChild (newNode (Italics),
	238	+ nodeAddChild (newNode (Bold), $2)); }
229	239	\| APO5 textnoboit APO3 textnoital APO2
230		~~- { debugf ("italicsorbold#10 "); $$ = nodeAddChild (newNode (Italics),~~
	240	+ { debugf ("italicsorbold#11 "); $$ = nodeAddChild (newNode (Italics),
231	241	makeTextBlock (nodeAddChild (newNode (Bold), $2), $4)); }
232		~~- \| APO5 textnoboit APO2 textnobold APO3~~
233		~~- { debugf ("italicsorbold#11 "); $$ = nodeAddChild (newNode (Bold),~~
234		~~- makeTextBlock (nodeAddChild (newNode (Italics), $2), $4)); }~~
235	242	\| APO5 textnoboit APO3 textnoital
236	243	{ debugf ("italicsorbold#12 "); $$ = makeTextBlock2 (newNodeS (TextToken, "''"),
237	244	nodeAddChild (newNode (Bold), $2), $4); }
	245	+ \| APO5 textnoboit APO3
	246	+ { debugf ("italicsorbold#13 "); $$ = makeTextBlock (newNodeS (TextToken, "''"),
	247	+ nodeAddChild (newNode (Bold), $2)); }
	248	+ \| APO5 textnoboit APO2 textnobold APO3
	249	+ { debugf ("italicsorbold#14 "); $$ = nodeAddChild (newNode (Bold),
	250	+ makeTextBlock (nodeAddChild (newNode (Italics), $2), $4)); }
238	251	\| APO5 textnoboit APO2 textnobold
239		~~- { debugf ("italicsorbold#13 "); $$ = makeTextBlock2 (newNodeS (TextToken, "'''"),~~
	252	+ { debugf ("italicsorbold#15 "); $$ = makeTextBlock2 (newNodeS (TextToken, "'''"),
240	253	nodeAddChild (newNode (Italics), $2), $4); }
	254	+ \| APO5 textnoboit APO2
	255	+ { debugf ("italicsorbold#16 "); $$ = makeTextBlock (newNodeS (TextToken, "'''"),
	256	+ nodeAddChild (newNode (Italics), $2)); }
241	257	\| APO5 textnoboit
242		~~- { debugf ("italicsorbold#14 ");~~
	258	+ { debugf ("italicsorbold#17 ");
243	259	$$ = makeTextBlock (newNodeS (TextToken, "'''''"), $2); }
244	260
245	261	italorboldnotbl : APO2 textnoitaltbl APO2
—	—	@@ -267,20 +283,29 @@
268	284	(TextToken, "'"), nodeAddChild (newNode (Italics), $2))); }
269	285	\| APO3 textnoboldtbl
270	286	{ debugf ("italorboldnotbl#9 "); $$ = makeTextBlock (newNodeS (TextToken, "'''"), $2); }
	287	+ \| APO5 textnoboittbl APO5
	288	+ { debugf ("italorboldnotbl#10 "); $$ = nodeAddChild (newNode (Italics),
	289	+ nodeAddChild (newNode (Bold), $2)); }
271	290	\| APO5 textnoboittbl APO3 textnoitaltbl APO2
272		~~- { debugf ("italorboldnotbl#10 "); $$ = nodeAddChild (newNode (Italics),~~
	291	+ { debugf ("italorboldnotbl#11 "); $$ = nodeAddChild (newNode (Italics),
273	292	makeTextBlock (nodeAddChild (newNode (Bold), $2), $4)); }
274		~~- \| APO5 textnoboittbl APO2 textnoboldtbl APO3~~
275		~~- { debugf ("italorboldnotbl#11 "); $$ = nodeAddChild (newNode (Bold),~~
276		~~- makeTextBlock (nodeAddChild (newNode (Italics), $2), $4)); }~~
277	293	\| APO5 textnoboittbl APO3 textnoitaltbl
278	294	{ debugf ("italorboldnotbl#12 "); $$ = makeTextBlock2 (newNodeS (TextToken, "''"),
279	295	nodeAddChild (newNode (Bold), $2), $4); }
	296	+ \| APO5 textnoboittbl APO3
	297	+ { debugf ("italorboldnotbl#13 "); $$ = makeTextBlock (newNodeS (TextToken, "''"),
	298	+ nodeAddChild (newNode (Bold), $2)); }
	299	+ \| APO5 textnoboittbl APO2 textnoboldtbl APO3
	300	+ { debugf ("italorboldnotbl#14 "); $$ = nodeAddChild (newNode (Bold),
	301	+ makeTextBlock (nodeAddChild (newNode (Italics), $2), $4)); }
280	302	\| APO5 textnoboittbl APO2 textnoboldtbl
281		~~- { debugf ("italorboldnotbl#13 "); $$ = makeTextBlock2 (newNodeS (TextToken, "'''"),~~
	303	+ { debugf ("italorboldnotbl#15 "); $$ = makeTextBlock2 (newNodeS (TextToken, "'''"),
282	304	nodeAddChild (newNode (Italics), $2), $4); }
	305	+ \| APO5 textnoboittbl APO2
	306	+ { debugf ("italorboldnotbl#16 "); $$ = makeTextBlock (newNodeS (TextToken, "'''"),
	307	+ nodeAddChild (newNode (Italics), $2)); }
283	308	\| APO5 textnoboittbl
284		~~- { debugf ("italorboldnotbl#14 ");~~
	309	+ { debugf ("italorboldnotbl#17 ");
285	310	$$ = makeTextBlock (newNodeS (TextToken, "'''''"), $2); }
286	311
287	312	italicsnobold : APO2 textnoboit APO2
—	—	@@ -303,20 +328,6 @@
304	329	\| APO3 textnoboittbl
305	330	{ debugf ("boldnoitalics#2 "); $$ = makeTextBlock (newNodeS (TextToken, "'''"), $2); }
306	331
307		~~-/* In order to resolve a reduce/reduce conflict correctly, heading must come before textelement. */~~
308		~~-heading : HEADING text ENDHEADING NEWLINE~~
309		~~- { debugf ("heading#1 "); $$ = nodeAddChild (newNodeI (Heading, $1), $2); }~~
310		~~- \| HEADING text ENDHEADING /* for eof */~~
311		~~- { debugf ("heading#2 "); $$ = nodeAddChild (newNodeI (Heading, $1), $2); }~~
312		~~- \| HEADING text NEWLINE~~
313		~~- { debugf ("heading#3 "); $$ = nodeAddChild (newNodeI (Heading, $1), $2); }~~
314		~~- \| HEADING text /* for eof */~~
315		~~- { debugf ("heading#4 "); $$ = nodeAddChild (newNodeI (Heading, $1), $2); }~~
316		~~- \| HEADING NEWLINE~~
317		~~- { debugf ("heading#5 "); $$ = nodeAddChild (newNodeI (Heading, $1), newNodeS (TextToken, "?")); }~~
318		~~- \| HEADING~~
319		~~- { debugf ("heading#6 "); $$ = nodeAddChild (newNodeI (Heading, $1), newNodeS (TextToken, "?")); }~~
320		-
321	332	table : TABLEBEGIN attributes tablerows TABLEEND
322	333	{ debugf ("table#1 "); $$ = nodeAddChild2 (newNode (Table), $2, $3); }
323	334	\| TABLEBEGIN attributes tablerows
—	—	@@ -418,7 +429,6 @@
419	430	textelement : TEXT { debugf ("textelement#1 "); $$ = $1; }
420	431	\| EXTENSION { debugf ("textelement#2 "); $$ = $1; }
421	432	\| PIPE { debugf ("textelement#3 "); $$ = newNodeS (TextToken, "\|"); }
422		~~- \| ENDHEADING { debugf ("textelement#4 "); $$ = processEndHeadingInText ($1); }~~
423	433	\| APO2 { debugf ("textelement#5 "); $$ = newNodeS (TextToken, "''"); }
424	434	\| APO3 { debugf ("textelement#6 "); $$ = newNodeS (TextToken, "'''"); }
425	435	\| APO5 { debugf ("textelement#7 "); $$ = newNodeS (TextToken, "'''''"); }
—	—	@@ -436,7 +446,6 @@
437	447	textelementnoital : TEXT { debugf ("textelementnoital#1 "); $$ = $1; }
438	448	\| EXTENSION { debugf ("textelementnoital#2 "); $$ = $1; }
439	449	\| PIPE { debugf ("textelementnoital#3 "); $$ = newNodeS (TextToken, "\|"); }
440		~~- \| ENDHEADING { debugf ("textelementnoital#4 "); $$ = processEndHeadingInText ($1); }~~
441	450	\| TABLEBEGIN { debugf ("textelementnoital#5 "); $$ = newNodeS (TextToken, addSpaces ("{\|", $1)); }
442	451	\| TABLEEND { debugf ("textelementnoital#6 "); $$ = newNodeS (TextToken, "\|}"); }
443	452	\| TABLEROW { debugf ("textelementnoital#7 "); $$ = convertTableRowToText ($1); }
—	—	@@ -449,7 +458,6 @@
450	459	textelementnobold : TEXT { debugf ("textelementnobold#1 "); $$ = $1; }
451	460	\| EXTENSION { debugf ("textelementnobold#2 "); $$ = $1; }
452	461	\| PIPE { debugf ("textelementnobold#3 "); $$ = newNodeS (TextToken, "\|"); }
453		~~- \| ENDHEADING { debugf ("textelementnobold#4 "); $$ = processEndHeadingInText ($1); }~~
454	462	\| TABLEBEGIN { debugf ("textelementnobold#5 "); $$ = newNodeS (TextToken, addSpaces ("{\|", $1)); }
455	463	\| TABLEEND { debugf ("textelementnobold#6 "); $$ = newNodeS (TextToken, "\|}"); }
456	464	\| TABLEROW { debugf ("textelementnobold#7 "); $$ = convertTableRowToText ($1); }
—	—	@@ -462,7 +470,6 @@
463	471	textelementnoboit : TEXT { debugf ("textelementnoboit#1 "); $$ = $1; }
464	472	\| EXTENSION { debugf ("textelementnoboit#2 "); $$ = $1; }
465	473	\| PIPE { debugf ("textelementnoboit#3 "); $$ = newNodeS (TextToken, "\|"); }
466		~~- \| ENDHEADING { debugf ("textelementnoboit#4 "); $$ = processEndHeadingInText ($1); }~~
467	474	\| TABLEBEGIN { debugf ("textelementnoboit#5 "); $$ = newNodeS (TextToken, addSpaces ("{\|", $1)); }
468	475	\| TABLEEND { debugf ("textelementnoboit#6 "); $$ = newNodeS (TextToken, "\|}"); }
469	476	\| TABLEROW { debugf ("textelementnoboit#7 "); $$ = convertTableRowToText ($1); }
—	—	@@ -474,7 +481,6 @@
475	482	textelementnotbl : TEXT { debugf ("textelementnotbl#1 "); $$ = $1; }
476	483	\| EXTENSION { debugf ("textelementnotbl#2 "); $$ = $1; }
477	484	\| PIPE { debugf ("textelementnotbl#3 "); $$ = newNodeS (TextToken, "\|"); }
478		~~- \| ENDHEADING { debugf ("textelementnotbl#4 "); $$ = processEndHeadingInText ($1); }~~
479	485	\| APO2 { debugf ("textelementnotbl#5 "); $$ = newNodeS (TextToken, "''"); }
480	486	\| APO3 { debugf ("textelementnotbl#6 "); $$ = newNodeS (TextToken, "'''"); }
481	487	\| APO5 { debugf ("textelementnotbl#7 "); $$ = newNodeS (TextToken, "'''''"); }
—	—	@@ -486,7 +492,6 @@
487	493	textelementnoitaltbl: TEXT { debugf ("textelementnoitaltbl#1 "); $$ = $1; }
488	494	\| EXTENSION { debugf ("textelementnoitaltbl#2 "); $$ = $1; }
489	495	\| PIPE { debugf ("textelementnoitaltbl#3 "); $$ = newNodeS (TextToken, "\|"); }
490		~~- \| ENDHEADING { debugf ("textelementnoitaltbl#4 "); $$ = processEndHeadingInText ($1); }~~
491	496	\| comment { debugf ("textelementnoitaltbl#5 "); $$ = $1; }
492	497	\| linketcnotbl { debugf ("textelementnoitaltbl#6 "); $$ = $1; }
493	498	\| boldnoitalicstbl{ debugf ("textelementnoitaltbl#7 "); $$ = $1; }
—	—	@@ -494,7 +499,6 @@
495	500	textelementnoboldtbl: TEXT { debugf ("textelementnoboldtbl#1 "); $$ = $1; }
496	501	\| EXTENSION { debugf ("textelementnoboldtbl#2 "); $$ = $1; }
497	502	\| PIPE { debugf ("textelementnoboldtbl#3 "); $$ = newNodeS (TextToken, "\|"); }
498		~~- \| ENDHEADING { debugf ("textelementnoboldtbl#4 "); $$ = processEndHeadingInText ($1); }~~
499	503	\| comment { debugf ("textelementnoboldtbl#5 "); $$ = $1; }
500	504	\| linketcnotbl { debugf ("textelementnoboldtbl#6 "); $$ = $1; }
501	505	\| italicsnoboldtbl{ debugf ("textelementnoboldtbl#7 "); $$ = $1; }
—	—	@@ -502,7 +506,6 @@
503	507	textelementnoboittbl: TEXT { debugf ("textelementnoboittbl#1 "); $$ = $1; }
504	508	\| EXTENSION { debugf ("textelementnoboittbl#2 "); $$ = $1; }
505	509	\| PIPE { debugf ("textelementnoboittbl#3 "); $$ = newNodeS (TextToken, "\|"); }
506		~~- \| ENDHEADING { debugf ("textelementnoboittbl#4 "); $$ = processEndHeadingInText ($1); }~~
507	510	\| comment { debugf ("textelementnoboittbl#5 "); $$ = $1; }
508	511	\| linketcnotbl { debugf ("textelementnoboittbl#6 "); $$ = $1; }
509	512
Index: trunk/flexbisonparse/wikilex.l
—	—	@@ -29,13 +29,28 @@
30	30	{
31	31	int i = 1;
32	32	while (input[i] == '-') i++;
33		~~- return ((i-1)*0x10000 + (initleng-i));~~
	33	+ return (((i-1) << 16) \| (initleng-i));
34	34	}
35	35
36	36	#define COMMONTOKENS \
37	37	if (YY_START != inattributeapo && YY_START != inattributeq && YY_START != canbeheading) \
38	38	BEGIN (cannotbelistorheadingorpre);
39	39
	40	+#define APOSTROPHE_CATASTROPHE(len,debugstr,token) \
	41	+ if (yytext[len] == '\'') \
	42	+ { \
	43	+ yyless (1); \
	44	+ yylval.node = newNodeS (TextToken, "'"); \
	45	+ debuglex2 ("TEXT(%s) ", "'"); \
	46	+ return TEXT; \
	47	+ } \
	48	+ else \
	49	+ { \
	50	+ yyless (len); \
	51	+ debuglex (debugstr); \
	52	+ return token; \
	53	+ }
	54	+
40	55	%}
41	56
42	57	/* inclusive start conditions */
—	—	@@ -122,9 +137,12 @@
123	138	\}\}\} { COMMONTOKENS; debuglex ("CLOSETEMPLATEVAR "); return CLOSETEMPLATEVAR; }
124	139	\{\{ { COMMONTOKENS; debuglex ("OPENTEMPLATE "); return OPENTEMPLATE; }
125	140	\}\} { COMMONTOKENS; debuglex ("CLOSETEMPLATE "); return CLOSETEMPLATE; }
126		~~-'''''/[^'] { COMMONTOKENS; debuglex ("APO5 "); return APO5; }~~
127		~~-'''/[^'] { COMMONTOKENS; debuglex ("APO3 "); return APO3; }~~
128		~~-''/[^'] { COMMONTOKENS; debuglex ("APO2 "); return APO2; }~~
	141	+''''' { COMMONTOKENS; debuglex ("APO5 "); return APO5; }
	142	+''' { COMMONTOKENS; debuglex ("APO3 "); return APO3; }
	143	+'' { COMMONTOKENS; debuglex ("APO2 "); return APO2; }
	144	+'''''. { COMMONTOKENS; APOSTROPHE_CATASTROPHE (5, "APO5 ", APO5); }
	145	+'''. { COMMONTOKENS; APOSTROPHE_CATASTROPHE (3, "APO3 ", APO3); }
	146	+''. { COMMONTOKENS; APOSTROPHE_CATASTROPHE (2, "APO2 ", APO2); }
129	147	\n { BEGIN (INITIAL); debuglex ("NEWLINE\n"); return NEWLINE; }
130	148	\r { /* ignore this one */ debuglex ("<13> "); }
131	149
—	—	@@ -140,10 +158,12 @@
141	159	debuglex2 ("HEADING(%d) ", yylval.num);
142	160	return HEADING;
143	161	}
144		~~-<canbeheading>"="+" "* {~~
	162	+<canbeheading>"="+" "*\r\n \|
	163	+<canbeheading>"="+" "*\n {
145	164	yylval.num = 0;
146	165	while (yytext [ yylval.num ] == '=')
147	166	yylval.num++;
	167	+ yylval.num \|= (yyleng-yylval.num) << 16;
148	168	debuglex2 ("ENDHEADING(%d) ", yylval.num);
149	169	return ENDHEADING;
150	170	}
Index: trunk/flexbisonparse/parsetree.c
—	—	@@ -295,14 +295,18 @@
296	296	Node processEndHeadingInText (int n)
297	297	{
298	298	char* ret;
	299	+ int equalses, spaces, i;
299	300
300		~~- if (n < 1) return 0;~~
301		~~- /* Performance optimisation */~~
302		~~- if (n == 1) return newNodeS (TextToken, "=");~~
	301	+ equalses = n % 0x10000;
	302	+ spaces = n >> 16;
303	303
304		~~- ret = (char) malloc ((n+1) sizeof (char));~~
305		~~- ret[n] = '\0';~~
306		~~- while (n) ret[--n] = '=';~~
	304	+ if (equalses + spaces < 1) return 0;
	305	+
	306	+ ret = (char) malloc ((equalses + spaces + 1) sizeof (char));
	307	+ i = 0;
	308	+ while (equalses--) ret[i++] = '=';
	309	+ while (spaces--) ret[i++] = ' ';
	310	+ ret[i] = '\0';
307	311	return newNodeS (TextToken, ret);
308	312	}
309	313
—	—	@@ -557,6 +561,18 @@
558	562	return newNodeS (TextToken, addSpaces (info % 2 ? "!" : "!!", info/2));
559	563	}
560	564
	565	+Node convertHeadingToText (int info)
	566	+{
	567	+ int i;
	568	+ char* text;
	569	+
	570	+ text = (char) malloc ((info + 1) sizeof (char));
	571	+ i = 0;
	572	+ while (info--) text[i++] = '=';
	573	+ text[i] = '\0';
	574	+ return newNodeS (TextToken, text);
	575	+}
	576	+
561	577	char* addSpaces (char* src, int spaces)
562	578	{
563	579	char* ret;

Status & tagging log

15:00, 12 September 2011 Meno25 (talk | contribs) changed the status of r5238 [removed: ok added: old]
13:39, 18 June 2009 😂 (talk | contribs) changed the status of r5238 [removed: new added: ok]