This patch (against 090800 sources) implements saving a document's
spellcheck ignored word list with the document when saved in the *.abw
format. The corresponding read back is also implemented. After this
patch, the "ignore all" word list acts like a persistent, per-document
custom dictionary. This patch includes the changes of my previous but
not-yet-committed patch on the same subject of a few days ago. In
other words, if someone decides to commit the attached patch, my
earlier patch can be discarded. If someone would rather see it a
different way, I could cook that up.
I have added two preference variables, SpellCheckIgnoredWordsSave and
SpellCheckIgnoredWordsLoad, and this time around they actually control
the feature (sorry, no GUI yet). As a side effect of the
implementation to use those preferences, PD_Document and AP_Convert
objects now have an XAP_App pointer (and a getApp()). The means that
lots of new places (including the import/export code) can now access
the user preferences. Go hog wild.
Ignored words are saved in a bit of XML that looks like this:
<ignoredwords>
<iw>sumthing</iw>
<iw>utherthing</iw>
</ignoredwords>
The <ignoredwords> tag is at the same level as <section> and <style>.
Most changes are XP, tested on Linux. I'm still not sure if this
works with the alternative Gnome XML parser, but my guess is that it
does. A small part of the changes is platform-specific, but things
should work if the code meets my modest assumptions about class
hierarchies and member functions. The Unix stuff works for sure.
--
[EMAIL PROTECTED] (WJCarpenter) PGP 0x91865119
38 95 1B 69 C9 C6 3D 25 73 46 32 04 69 D6 ED F3
Buy my house in Woodinville (near Seattle):
<http://www.johnlscott.com/57554>
diff -ru abi-090800-ORIG/src/text/ptbl/xp/pd_Document.cpp
abi-090800/src/text/ptbl/xp/pd_Document.cpp
--- abi-090800-ORIG/src/text/ptbl/xp/pd_Document.cpp Thu Jul 6 01:55:57 2000
+++ abi-090800/src/text/ptbl/xp/pd_Document.cpp Sat Sep 9 22:28:55 2000
@@ -50,9 +50,10 @@
//////////////////////////////////////////////////////////////////
//////////////////////////////////////////////////////////////////
-PD_Document::PD_Document()
+PD_Document::PD_Document(XAP_App *pApp)
: AD_Document(), m_hashDataItems(11)
{
+ m_pApp = pApp;
m_pPieceTable = NULL;
// perhaps this should be a magic "unknown" or "NULL" value,
diff -ru abi-090800-ORIG/src/text/ptbl/xp/pd_Document.h
abi-090800/src/text/ptbl/xp/pd_Document.h
--- abi-090800-ORIG/src/text/ptbl/xp/pd_Document.h Wed Jul 26 21:15:26 2000
+++ abi-090800/src/text/ptbl/xp/pd_Document.h Sat Sep 9 22:29:43 2000
@@ -44,6 +44,7 @@
class pf_Frag_Strux;
class PX_ChangeRecord;
class PD_Style;
+class XAP_App;
#ifdef PT_TEST
#include "ut_test.h"
@@ -61,7 +62,7 @@
class PD_Document : public AD_Document
{
public:
- PD_Document();
+ PD_Document(XAP_App *pApp);
virtual UT_Error readFromFile(const char * szFilename, int
ieft);
virtual UT_Error newDocument(void);
@@ -168,6 +169,7 @@
const char * getFileName() { return m_szFilename; }
UT_uint32 getLastType() { return
m_lastSavedAsType; }
+ XAP_App * getApp() { return m_pApp; }
#ifdef PT_TEST
void __dump(FILE * fp) const;
@@ -185,6 +187,7 @@
UT_AlphaHashTable m_hashDataItems;
IEFileType m_lastSavedAsType;
+ XAP_App * m_pApp;
};
diff -ru abi-090800-ORIG/src/wp/ap/beos/ap_BeOSApp.cpp
abi-090800/src/wp/ap/beos/ap_BeOSApp.cpp
--- abi-090800-ORIG/src/wp/ap/beos/ap_BeOSApp.cpp Wed Aug 16 13:35:41 2000
+++ abi-090800/src/wp/ap/beos/ap_BeOSApp.cpp Sun Sep 10 10:47:54 2000
@@ -666,7 +666,7 @@
#if CONVERT
if (to)
{
- AP_Convert * conv = new AP_Convert();
+ AP_Convert * conv = new AP_Convert(getApp());
conv->setVerbose(verbose);
conv->convertTo(m_pArgs->m_argv[k], to);
delete conv;
diff -ru abi-090800-ORIG/src/wp/ap/beos/ap_BeOSFrame.cpp
abi-090800/src/wp/ap/beos/ap_BeOSFrame.cpp
--- abi-090800-ORIG/src/wp/ap/beos/ap_BeOSFrame.cpp Sat Aug 19 16:35:39 2000
+++ abi-090800/src/wp/ap/beos/ap_BeOSFrame.cpp Sun Sep 10 10:27:56 2000
@@ -416,7 +416,7 @@
// load a document into the current frame.
// if no filename, create a new document.
- AD_Document * pNewDoc = new PD_Document();
+ AD_Document * pNewDoc = new PD_Document(getApp());
UT_ASSERT(pNewDoc);
if (!szFilename || !*szFilename)
diff -ru abi-090800-ORIG/src/wp/ap/qnx/ap_QNXFrame.cpp
abi-090800/src/wp/ap/qnx/ap_QNXFrame.cpp
--- abi-090800-ORIG/src/wp/ap/qnx/ap_QNXFrame.cpp Tue Aug 15 09:20:27 2000
+++ abi-090800/src/wp/ap/qnx/ap_QNXFrame.cpp Sun Sep 10 10:28:08 2000
@@ -463,7 +463,7 @@
// load a document into the current frame.
// if no filename, create a new document.
- AD_Document * pNewDoc = new PD_Document();
+ AD_Document * pNewDoc = new PD_Document(getApp());
UT_ASSERT(pNewDoc);
if (!szFilename || !*szFilename)
diff -ru abi-090800-ORIG/src/wp/ap/unix/ap_UnixApp.cpp
abi-090800/src/wp/ap/unix/ap_UnixApp.cpp
--- abi-090800-ORIG/src/wp/ap/unix/ap_UnixApp.cpp Sun Aug 27 18:45:41 2000
+++ abi-090800/src/wp/ap/unix/ap_UnixApp.cpp Sun Sep 10 10:47:07 2000
@@ -1011,7 +1011,7 @@
// [filename]
if (to)
{
- AP_Convert * conv = new AP_Convert();
+ AP_Convert * conv = new AP_Convert(getApp());
conv->setVerbose(verbose);
conv->convertTo(m_pArgs->m_argv[k], to);
delete conv;
diff -ru abi-090800-ORIG/src/wp/ap/unix/ap_UnixFrame.cpp
abi-090800/src/wp/ap/unix/ap_UnixFrame.cpp
--- abi-090800-ORIG/src/wp/ap/unix/ap_UnixFrame.cpp Thu Jul 27 20:24:11 2000
+++ abi-090800/src/wp/ap/unix/ap_UnixFrame.cpp Sun Sep 10 10:28:20 2000
@@ -425,7 +425,7 @@
// load a document into the current frame.
// if no filename, create a new document.
- AD_Document * pNewDoc = new PD_Document();
+ AD_Document * pNewDoc = new PD_Document(getApp());
UT_ASSERT(pNewDoc);
if (!szFilename || !*szFilename)
diff -ru abi-090800-ORIG/src/wp/ap/unix/gnome/ap_UnixGnomeApp.cpp
abi-090800/src/wp/ap/unix/gnome/ap_UnixGnomeApp.cpp
--- abi-090800-ORIG/src/wp/ap/unix/gnome/ap_UnixGnomeApp.cpp Sun Aug 27 19:41:01
2000
+++ abi-090800/src/wp/ap/unix/gnome/ap_UnixGnomeApp.cpp Sun Sep 10 10:48:08 2000
@@ -277,7 +277,7 @@
}
if (to) {
- AP_Convert * conv = new AP_Convert();
+ AP_Convert * conv = new AP_Convert(getApp());
conv->setVerbose(verbose);
while ((file = poptGetArg (poptcon)) != NULL) {
diff -ru abi-090800-ORIG/src/wp/ap/win/ap_Win32App.cpp
abi-090800/src/wp/ap/win/ap_Win32App.cpp
--- abi-090800-ORIG/src/wp/ap/win/ap_Win32App.cpp Mon Jul 10 14:56:16 2000
+++ abi-090800/src/wp/ap/win/ap_Win32App.cpp Sun Sep 10 10:48:18 2000
@@ -867,7 +867,7 @@
// [filename]
if (to)
{
- AP_Convert * conv = new AP_Convert();
+ AP_Convert * conv = new AP_Convert(getApp());
conv->setVerbose(verbose);
conv->convertTo(m_pArgs->m_argv[k], to);
delete conv;
diff -ru abi-090800-ORIG/src/wp/ap/win/ap_Win32Frame.cpp
abi-090800/src/wp/ap/win/ap_Win32Frame.cpp
--- abi-090800-ORIG/src/wp/ap/win/ap_Win32Frame.cpp Tue Sep 5 02:01:32 2000
+++ abi-090800/src/wp/ap/win/ap_Win32Frame.cpp Sun Sep 10 10:28:42 2000
@@ -1112,7 +1112,7 @@
// load a document into the current frame.
// if no filename, create a new document.
- AD_Document * pNewDoc = new PD_Document();
+ AD_Document * pNewDoc = new PD_Document(getApp());
UT_ASSERT(pNewDoc);
if (!szFilename || !*szFilename)
diff -ru abi-090800-ORIG/src/wp/ap/xp/ap_Convert.cpp
abi-090800/src/wp/ap/xp/ap_Convert.cpp
--- abi-090800-ORIG/src/wp/ap/xp/ap_Convert.cpp Wed Mar 8 14:35:15 2000
+++ abi-090800/src/wp/ap/xp/ap_Convert.cpp Sun Sep 10 10:44:46 2000
@@ -24,11 +24,14 @@
#include "ie_exp.h"
#include "ut_types.h"
+class XAP_App;
+
//////////////////////////////////////////////////////////////////
-AP_Convert::AP_Convert(void)
+AP_Convert::AP_Convert(XAP_App *pApp)
{
m_iVerbose = 1;
+ m_pApp = pApp;
}
AP_Convert::~AP_Convert(void)
@@ -42,7 +45,7 @@
const char * szTargetFilename,
IEFileType targetFormat)
{
- PD_Document * pNewDoc = new PD_Document();
+ PD_Document * pNewDoc = new PD_Document(getApp());
UT_Error error;
UT_ASSERT(pNewDoc);
diff -ru abi-090800-ORIG/src/wp/ap/xp/ap_Convert.h abi-090800/src/wp/ap/xp/ap_Convert.h
--- abi-090800-ORIG/src/wp/ap/xp/ap_Convert.h Wed Mar 8 14:35:15 2000
+++ abi-090800/src/wp/ap/xp/ap_Convert.h Sun Sep 10 10:45:16 2000
@@ -28,7 +28,7 @@
class AP_Convert
{
public:
- AP_Convert(void);
+ AP_Convert(XAP_App *pApp);
~AP_Convert(void);
void convertTo(const char * szSourceFilename,
@@ -44,9 +44,11 @@
const char * szTargetFormat);
void setVerbose(int level);
+ XAP_App* getApp() { return m_pApp; }
private:
int m_iVerbose;
+ XAP_App *m_pApp;
};
#endif /* AP_CONVERT_H */
diff -ru abi-090800-ORIG/src/wp/ap/xp/ap_Prefs_SchemeIds.h
abi-090800/src/wp/ap/xp/ap_Prefs_SchemeIds.h
--- abi-090800-ORIG/src/wp/ap/xp/ap_Prefs_SchemeIds.h Tue Jul 25 14:10:51 2000
+++ abi-090800/src/wp/ap/xp/ap_Prefs_SchemeIds.h Sun Sep 10 11:14:47 2000
@@ -55,6 +55,12 @@
#define AP_PREF_KEY_SpellCheckInternet "SpellCheckInternet"
/* enable spell checking internet names {0,1} */
#define AP_PREF_DEFAULT_SpellCheckInternet "1"
+#define AP_PREF_KEY_SpellCheckIgnoredWordsSave "SpellCheckIgnoredWordsSave"
+ /* save ignored words list with doc */
+#define AP_PREF_DEFAULT_SpellCheckIgnoredWordsSave "1"
+
+#define AP_PREF_KEY_SpellCheckIgnoredWordsLoad "SpellCheckIgnoredWordsLoad"
+ /* load ignored words list with doc */
+#define AP_PREF_DEFAULT_SpellCheckIgnoredWordsLoad "1"
+
#define AP_PREF_KEY_OptionsTabNumber "OptionsTabNumber"
/* the page number of the currently shown page in the */
#define AP_PREF_DEFAULT_OptionsTabNumber "0"
/* options dialog */
@@ -122,6 +128,8 @@
dcl(SpellCheckCaps)
dcl(SpellCheckNumbers)
dcl(SpellCheckInternet)
+dcl(SpellCheckIgnoredWordsSave)
+dcl(SpellCheckIgnoredWordsLoad)
dcl(OptionsTabNumber)
dcl(RulerUnits)
dcl(RulerVisible)
diff -ru abi-090800-ORIG/src/wp/impexp/xp/ie_exp_AbiWord_1.cpp
abi-090800/src/wp/impexp/xp/ie_exp_AbiWord_1.cpp
--- abi-090800-ORIG/src/wp/impexp/xp/ie_exp_AbiWord_1.cpp Wed Jul 26 21:15:31
2000
+++ abi-090800/src/wp/impexp/xp/ie_exp_AbiWord_1.cpp Sun Sep 10 11:36:46 2000
@@ -21,6 +21,7 @@
#include "ut_types.h"
#include "ut_bytebuf.h"
#include "ut_base64.h"
+#include "ut_debugmsg.h"
#include "pt_Types.h"
#include "ie_exp_AbiWord_1.h"
#include "pd_Document.h"
@@ -30,6 +31,7 @@
#include "px_CR_Span.h"
#include "px_CR_Strux.h"
#include "xap_App.h"
+#include "ap_Prefs.h"
#include "pd_Style.h"
/*****************************************************************/
@@ -116,6 +118,7 @@
UT_Bool
bNewLineAfter, PT_AttrPropIndex api);
void _outputData(const UT_UCSChar * p, UT_uint32
length);
void _handleStyles(void);
+ void _handleIgnoredWords(void);
void _handleDataItems(void);
PD_Document * m_pDocument;
@@ -455,6 +458,7 @@
_handleStyles();
+ _handleIgnoredWords();
}
s_AbiWord_1_Listener::~s_AbiWord_1_Listener()
@@ -620,6 +624,61 @@
if (bWroteOpenStyleSection)
m_pie->write("</styles>\n");
+
+ return;
+}
+
+void s_AbiWord_1_Listener::_handleIgnoredWords(void)
+{
+ UT_ASSERT(m_pDocument);
+ XAP_App *pApp = m_pDocument->getApp();
+ UT_ASSERT(pApp);
+ XAP_Prefs *pPrefs = pApp->getPrefs();
+ UT_ASSERT(pPrefs);
+
+ UT_Bool saveIgnores;
+ pPrefs->getPrefsValueBool((XML_Char *)AP_PREF_KEY_SpellCheckIgnoredWordsSave,
+&saveIgnores);
+ UT_DEBUGMSG(("Ignored words list %s being saved with document\n",
+saveIgnores?"is":"is not"));
+ if (!saveIgnores) return; // don't bother
+ UT_Bool bWroteOpenIgnoredWordsSection = UT_FALSE;
+
+ const UT_UCSChar *word;
+ for (UT_uint32 i = 0; m_pDocument->enumIgnores(i, &word); i++)
+ {
+ if (!bWroteOpenIgnoredWordsSection)
+ {
+ m_pie->write("<ignoredwords>\n");
+ bWroteOpenIgnoredWordsSection = UT_TRUE;
+ }
+ m_pie->write("<iw>");
+ for (UT_uint32 udex=0; word[udex]; ++udex)
+ {
+ UT_UCSChar ch = word[udex];
+ switch (ch)
+ {
+ case '&': m_pie->write("&"); break;
+ case '<': m_pie->write("<"); break;
+ case '>': m_pie->write(">"); break;
+ case '"': m_pie->write("""); break;
+ default:
+ char utb[100];
+ if (ch < ' ' || ch >= 128)
+ {
+ sprintf(utb, "&#x%x;", ch);
+ }
+ else
+ {
+ utb[0] = (char)ch;
+ utb[1] = 0;
+ }
+ m_pie->write(utb);
+ }
+ }
+ m_pie->write("</iw>\n");
+ }
+
+ if (bWroteOpenIgnoredWordsSection)
+ m_pie->write("</ignoredwords>\n");
return;
}
diff -ru abi-090800-ORIG/src/wp/impexp/xp/ie_imp_AbiWord_1.cpp
abi-090800/src/wp/impexp/xp/ie_imp_AbiWord_1.cpp
--- abi-090800-ORIG/src/wp/impexp/xp/ie_imp_AbiWord_1.cpp Wed Jul 26 21:15:31
2000
+++ abi-090800/src/wp/impexp/xp/ie_imp_AbiWord_1.cpp Sun Sep 10 11:49:28 2000
@@ -33,6 +33,8 @@
#include "ie_types.h"
#include "pd_Document.h"
#include "ut_bytebuf.h"
+#include "xap_Prefs.h"
+#include "ap_Prefs.h"
/*****************************************************************
******************************************************************
@@ -259,6 +261,8 @@
#define TT_PAGEBREAK 11 // a forced page-break <pbr>
#define TT_STYLESECTION 12 // a style section <styles>
#define TT_STYLE 13 // a style <s> within a style section
+#define TT_IGNOREDWORDS 14 // an ignored words section <ignoredwords>
+#define TT_IGNOREDWORD 15 // a word <iw> within an ignored words section
struct _TokenTable
{
@@ -297,6 +301,8 @@
{ "pbr", TT_PAGEBREAK },
{ "styles", TT_STYLESECTION },
{ "s", TT_STYLE },
+ { "ignoredwords", TT_IGNOREDWORDS },
+ { "iw", TT_IGNOREDWORD },
{ "*", TT_OTHER }}; // must be last
#define TokenTableSize ((sizeof(s_Tokens)/sizeof(s_Tokens[0])))
@@ -339,6 +345,12 @@
xxx_UT_DEBUGMSG(("startElement: %s\n", name));
X_EatIfAlreadyError(); // xml parser keeps running
until buffer consumed
+
+ UT_ASSERT(m_pDocument);
+ XAP_App *pApp = m_pDocument->getApp();
+ UT_ASSERT(pApp);
+ XAP_Prefs *pPrefs = pApp->getPrefs();
+ UT_ASSERT(pPrefs);
UT_uint32 tokenIndex = s_mapNameToToken(name);
switch (s_Tokens[tokenIndex].m_type)
@@ -456,6 +468,23 @@
X_CheckError(m_pDocument->appendStyle(atts));
return;
+ case TT_IGNOREDWORDS:
+ X_VerifyParseState(_PS_Doc);
+ // This caches the preference value. Our assumption is that the
+ignored words
+ // list is small with respect to the document size, but nothing forces
+that.
+ // The scheme is to parse the ignored words list as usual, but if we
+don't want
+ // it loaded from the file, it just isn't added to the in-memory
+ignored words
+ // list. The cached preference value keeps us from looking it up for
+each word.
+ pPrefs->getPrefsValueBool((XML_Char
+*)AP_PREF_KEY_SpellCheckIgnoredWordsLoad, &m_bLoadIgnoredWords);
+
+ m_parseState = _PS_IgnoredWordsSec;
+ return;
+
+ case TT_IGNOREDWORD:
+ X_VerifyParseState(_PS_IgnoredWordsSec);
+ m_parseState = _PS_IgnoredWordsItem;
+ return;
+
case TT_OTHER:
default:
UT_DEBUGMSG(("Unknown tag [%s]\n",name));
@@ -564,6 +593,16 @@
m_parseState = _PS_StyleSec;
return;
+ case TT_IGNOREDWORDS:
+ X_VerifyParseState(_PS_IgnoredWordsSec);
+ m_parseState = _PS_Doc;
+ return;
+
+ case TT_IGNOREDWORD:
+ X_VerifyParseState(_PS_IgnoredWordsItem);
+ m_parseState = _PS_IgnoredWordsSec;
+ return;
+
case TT_OTHER:
default:
UT_DEBUGMSG(("Unknown end tag [%s]\n",name));
@@ -591,6 +630,7 @@
}
case _PS_Block:
+ case _PS_IgnoredWordsItem:
{
UT_ASSERT(sizeof(XML_Char) == sizeof(UT_Byte));
UT_ASSERT(sizeof(XML_Char) != sizeof(UT_UCSChar));
@@ -600,19 +640,17 @@
// [] convert CRLF to SP.
// [] convert CR to SP.
// [] convert LF to SP.
+ // ignored words processing doesn't care about the
+ // white-space stuff, but it does no harm
UT_Byte * ss = (UT_Byte *)s;
- UT_UCSChar buf[1024];
+ UT_UCSChar _buf[1024], *buf = _buf;
+ // len is an upper bound on the length of the decoded stuff
+ if (len > 1000) buf = new UT_UCSChar[len+1];
int bufLen = 0;
for (int k=0; k<len; k++)
{
- if (bufLen == NrElements(buf)) // pump it out
in chunks
- {
-
X_CheckError(m_pDocument->appendSpan(buf,bufLen));
- bufLen = 0;
- }
-
if ((ss[k] < 0x80) && (m_lenCharDataSeen > 0))
{
// is it us-ascii and we are in a UTF-8
@@ -674,10 +712,24 @@
}
}
- // flush out the last piece of a buffer
+ // flush out the buffer
if (bufLen > 0)
- X_CheckError(m_pDocument->appendSpan(buf,bufLen));
+ {
+ switch (m_parseState)
+ {
+ case _PS_Block:
+
+X_CheckError(m_pDocument->appendSpan(buf,bufLen));
+ break;
+ case _PS_IgnoredWordsItem:
+ if (m_bLoadIgnoredWords)
+X_CheckError(m_pDocument->appendIgnore(buf,bufLen));
+ break;
+ default:
+ UT_ASSERT(UT_SHOULD_NOT_HAPPEN);
+ break;
+ }
+ }
+ if (buf != _buf) delete buf;
return;
}
diff -ru abi-090800-ORIG/src/wp/impexp/xp/ie_imp_AbiWord_1.h
abi-090800/src/wp/impexp/xp/ie_imp_AbiWord_1.h
--- abi-090800-ORIG/src/wp/impexp/xp/ie_imp_AbiWord_1.h Wed Jul 26 21:15:31 2000
+++ abi-090800/src/wp/impexp/xp/ie_imp_AbiWord_1.h Sun Sep 10 11:41:42 2000
@@ -82,11 +82,14 @@
_PS_DataSec,
_PS_DataItem,
_PS_StyleSec,
- _PS_Style
+ _PS_Style,
+ _PS_IgnoredWordsSec,
+ _PS_IgnoredWordsItem
} ParseState;
UT_Error m_error;
ParseState m_parseState;
+ UT_Bool m_bLoadIgnoredWords;
XML_Char m_charDataSeen[4];
UT_uint32 m_lenCharDataSeen;
UT_uint32 m_lenCharDataExpected;
