/* ** $VER: Funcs.c 1.0 ** ** Library functions for HTML datatype */ #define __USE_SYSBASE #include #include #include #include #include #include "ixdt.h" /* ** Every iX Datatype has three functions: ** ** DTF_Identify() which performs some tests on given file ** to determine if that file should be read by this datatype. ** Function returns an integer value between 0 and 9. ** Value of 9 means that file is recogized 100%. ** Value of 0 means that file is not recognized. ** Value of e.g. 5 means that datatype is about 50% sure that ** file is of required type. ** Value of e.g. 8 means that datatype is almost 100% sure that ** file is of required type etc. ** ** DTF_FileStruct() which is used to determine begin-end of ** some logical parts of file. ** For example, AmigaGuide database is made up of nodes, and this ** function should find begin, end and name of every node. ** ** DTF_ProcessChunk() which reads one part of file and returns ** information on that part in iX_ChunkInfo form readable by iX-Guide */ long ec,ecc,ParserECS; ULONG __saveds __asm DTF_Identify(register __a0 STRPTR name) { /* ** we search for a tag but if it is not first tag or ** if it is not first after we return 0 which ** means it is not HTML. */ char buff[1],hbuff[5]; BPTR fh; if (fh=Open(name,MODE_OLDFILE)) { while(FRead(fh,buff,1,1)) { if (*buff=='<') { FRead(fh,hbuff,1,5); if (strnicmp(hbuff,"HTML>",5)==0) {Close(fh);return(9L);}; if (*hbuff=='!') { while(FRead(fh,buff,1,1)) if (*buff=='>') break; } else break; } else if ((*buff!=' ') && (*buff!='\n')) break; }; Close(fh); }; return(0L); } struct iX_Node * __saveds __asm DTF_FileStruct(register __a0 BPTR fh, register __d0 long from, register __a1 BOOL *save_index) { struct iX_Node *nd=NULL; *save_index = FALSE; // we don't want iX-Guide to save node index // since we have only one node if (!from) // if from is 0 which means we are at the begining // of file, we build our node, otherwise we are called // for a second time and we return NULL since we have // only one node { nd = AllocVec(sizeof(struct iX_Node),MEMF_CLEAR); nd->name = AllocVec(5,0); strcpy(nd->name,"main"); // "main" node will be loaded first nd->node_begin=0L; // node begin is actually begining of file Seek(fh,0L,OFFSET_END); nd->node_end=Seek(fh,0L,OFFSET_CURRENT); // node end is end of file }; return(nd); } /************ support functions ************/ void *Realloc(void *oldptr,long newsz,long oldsz) { void *mempt; mempt = AllocVec(newsz,MEMF_ANY); CopyMem(oldptr,mempt,oldsz); FreeVec(oldptr); return(mempt); } BOOL CheckMemory(struct iX_ChunkInfo *tpp,long *chunk_c,long entry_c,long ParserECS) { if (*chunk_c==ParserECS) { *chunk_c=0; tpp->ci_entry=Realloc(tpp->ci_entry,entry_c+ParserECS,entry_c); tpp->cie_position=Realloc(tpp->cie_position,4*entry_c+ParserECS*4,4*entry_c); tpp->cie_length=Realloc(tpp->cie_length,2*entry_c+ParserECS*2,2*entry_c); }; if (!tpp->ci_entry || !tpp->cie_position || !tpp->cie_length) return(FALSE); } void InsertEntries(struct iX_ChunkInfo *tpp,UBYTE *array,long num_e) { long ecn; for(ecn=0;ecnci_entry[ec] = array[ecn]; ec++;ecc++; CheckMemory(tpp,&ecc,ec,ParserECS); }; } long FindN(char *t,char **arr) { long nnum=0; while(arr[nnum]) { if (stricmp(t,arr[nnum])==0) break; nnum++; }; return(nnum+1); } /********************************/ /*** parser stuff ***/ #define TYPE_TAG 1 #define TYPE_WORD 2 #define TYPE_RTRN 3 #define TYPE_SPC 4 /*** HTML tags ***/ #define TAG_A 1 #define TAG_TITLE 2 #define TAG_B 3 #define TAG_I 4 #define TAG_U 5 #define TAG_H1 6 #define TAG_H2 7 #define TAG_H3 8 #define TAG_H4 9 #define TAG_H5 10 #define TAG_H6 11 #define TAG_CENTER 12 #define TAG_P 13 #define TAG_BR 14 #define TAG_OL 15 #define TAG_UL 16 #define TAG_LI 17 #define TAG_IMG 18 struct iX_ChunkInfo * __saveds __asm DTF_ProcessChunk(register __a0 STRPTR name, register __d0 long from, register __d1 long to, register __d2 BOOL head) { struct iX_ChunkInfo *ci=NULL; BPTR fhan; char *author="Created by HTML datatype V12.27"; // this will show up // in iXG INFO window char *deffont="F1"; char *dp,store,*beg; UBYTE header[]={ ET_STAG,ET_DATABASE,ET_ATTR,ET_RTRN, // @database tag ET_STAG,ET_AUTHOR,ET_ATTR,ET_RTRN, // @author tag ET_STAG,ET_FONT,ET_ATTR,ET_RTRN, ET_STAG,ET_IXF,ET_RTRN,ET_END}; // @ixf tag (enable iXG formatting) UBYTE link[]={ET_ATTR,ET_LINK,ET_ATTR}; UBYTE fontspec[]={ET_STAG,ET_FONT,ET_ATTR,ET_RTRN}; UBYTE align[]={ET_A,ET_ATTR,ET_ATTR}; UBYTE img[]={ET_ATTR,ET_PIC}; char *HTMLtags[]={"a","title","b","i","u","h1","h2","h3","h4", "h5","h6","center","p","br","ol","ul","li", "img",NULL}; char *ixg_strings = "F0F1F2F3F4F5F6DEFFLCRDWUPDEFA"; // fonts alignement long nn,etype; BOOL neg,hreffnd,inlink; if (head) // iX-Guide wants header of database { ci = AllocVec(sizeof(struct iX_ChunkInfo),MEMF_CLEAR); ci->ci_data = AllocVec(strlen(author)+strlen(name)+strlen(deffont)+2,MEMF_CLEAR); strcpy(ci->ci_data,author);strcat(ci->ci_data,name);strcat(ci->ci_data,deffont); ci->ci_datalen=strlen(author)+strlen(name)+strlen(deffont)+2; ci->ci_entrynum=16; ci->ci_entry=AllocVec(16,MEMF_CLEAR); ec=0;ecc=0; InsertEntries(ci,header,16); ci->cie_position=AllocVec(16*4,MEMF_CLEAR); ci->cie_position[2] = strlen(author);ci->cie_position[6] = 0L; ci->cie_position[10] = strlen(author)+strlen(name); ci->cie_length=AllocVec(16*2,MEMF_CLEAR); ci->cie_length[2] = strlen(name); ci->cie_length[6]=strlen(author); ci->cie_length[10] = strlen(deffont); } else // process chunk { ParserECS=1000; if ((to-from) > 50000) ParserECS=5000; else if ((to-from) > 20000) ParserECS=3000; ci = AllocVec(sizeof(struct iX_ChunkInfo),MEMF_CLEAR); ci->ci_entry = AllocVec(ParserECS,MEMF_ANY); ci->cie_position = AllocVec(ParserECS*4,MEMF_ANY); ci->cie_length = AllocVec(ParserECS*2,MEMF_ANY); ci->ci_data = AllocVec(to-from+2+strlen(ixg_strings),MEMF_ANY); if (!ci || !ci->ci_entry || !ci->cie_position || !ci->cie_length || !ci->ci_data) return(NULL); if (fhan=Open(name,MODE_OLDFILE)) { ci->ci_datalen=strlen(ixg_strings); strcpy(ci->ci_data,ixg_strings); Seek(fhan,from,OFFSET_BEGINING); ci->ci_datalen += FRead(fhan,ci->ci_data+strlen(ixg_strings),1,to-from+1); ci->ci_data[ci->ci_datalen] = 0; dp = ci->ci_data+strlen(ixg_strings); ec=0;ecc=0; while(*dp) { etype=0; switch(*dp) { case '<': etype=TYPE_TAG; break; case ' ': etype=TYPE_SPC; break; case '\n': etype=TYPE_RTRN; break; default: etype=TYPE_WORD; break; }; switch(etype) { case TYPE_WORD: ci->ci_entry[ec]=ET_WORD; ci->cie_position[ec]=dp-ci->ci_data; ci->cie_length[ec]=0L; while(*dp) { if (*dp==' ') {dp++;ci->cie_length[ec]++;break;}; if (*dp=='\n') break; if (*dp=='<') break; dp++;ci->cie_length[ec]++; }; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TYPE_SPC: ci->ci_entry[ec]=ET_SPC; ci->cie_length[ec]=0L; while(*dp && (*dp==' ')) {dp++;ci->cie_length[ec]++;}; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TYPE_RTRN: ci->ci_entry[ec]=ET_RTRN; *dp=' '; dp++; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TYPE_TAG: ci->ci_entry[ec]=ET_TAG; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); dp++; if (*dp=='!') while(*dp && (*dp!='>')) dp++; // comment else { if (*dp=='/') {neg=TRUE;dp++;} else neg=FALSE; while(*dp==' ') dp++; beg=dp; while(*dp && (*dp!='>') && (*dp!=' ')) dp++; store=*dp; *dp='\0'; nn = FindN(beg,HTMLtags); *dp=store; switch(nn) { case TAG_A: if (neg) inlink=FALSE; hreffnd=FALSE; while (*dp && (*dp!='>')) { while(*dp==' ') dp++; if (strnicmp(dp,"HREF=",5)==0) { hreffnd=TRUE;inlink=TRUE; dp+=5; if(*dp=='"') { dp++; if (strnicmp(dp,"file://localhost/",17)==0) dp+=17; else if (strnicmp(dp,"file:///",8)==0) dp+=8; beg=dp; while(*dp && (*dp!='"')) dp++; } else { if (strnicmp(dp,"file://localhost/",17)==0) dp+=17; else if (strnicmp(dp,"file:///",8)==0) dp+=8; beg=dp; while(*dp && (*dp!=' ') && (*dp!='>')) dp++; }; store=*dp;*dp='\0'; strcpy(beg-5,beg); if (!strchr(beg-5,'#')) strcat(beg-5,"/main"); InsertEntries(ci,link,3); ci->cie_position[ec-1]=beg-ci->ci_data-5; ci->cie_length[ec-1]=strlen(beg-5); ci->cie_position[ec-3]=beg-ci->ci_data-5; ci->cie_length[ec-3]=strlen(beg-5); *dp=store; while(*dp && (*dp!='>')) dp++; if (*dp=='>') dp++; if (*dp!='<') { ci->cie_position[ec-3]=dp-ci->ci_data; ci->cie_length[ec-3]=0; while(*dp && (*dp!='<')) {dp++;ci->cie_length[ec-3]++;}; }; break; } else while(*dp && (*dp!='>') && (*dp!=' ')) dp++; }; break; case TAG_IMG: while (*dp && (*dp!='>')) { while(*dp==' ') dp++; if (strnicmp(dp,"SRC=",4)==0) { dp+=4; if(*dp=='"') { dp++; if (strnicmp(dp,"file://localhost/",17)==0) dp+=17; else if (strnicmp(dp,"file:///",8)==0) dp+=8; beg=dp; while(*dp && (*dp!='"')) dp++; } else { if (strnicmp(dp,"file://localhost/",17)==0) dp+=17; else if (strnicmp(dp,"file:///",8)==0) dp+=8; beg=dp; while(*dp && (*dp!=' ') && (*dp!='>')) dp++; }; store=*dp;*dp='\0'; InsertEntries(ci,img,2); ci->cie_position[ec-2]=beg-ci->ci_data; ci->cie_length[ec-2]=strlen(beg); *dp=store; break; } else while(*dp && (*dp!='>') && (*dp!=' ')) dp++; }; break; case TAG_TITLE: ci->ci_entry[ec-1]=ET_RTRN; InsertEntries(ci,fontspec,4); if (neg) ci->cie_position[ec-2]=2; // FONT F1 if tag is else ci->cie_position[ec-2]=10; // FONT 5 in ixg_strings ci->cie_length[ec-2]=2; break; case TAG_H1: case TAG_H2: case TAG_H3: case TAG_H4: case TAG_H5: case TAG_H6: ci->ci_entry[ec-1]=ET_RTRN; InsertEntries(ci,fontspec,4); if (neg) ci->cie_position[ec-2]=2; // F1 if else ci->cie_position[ec-2]=12-(nn-6)*2; // H1 is F6, H2 is F5... ci->cie_length[ec-2]=2; break; case TAG_B: if (neg) ci->ci_entry[ec]=ET_UB; else ci->ci_entry[ec]=ET_B; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TAG_I: if (neg) ci->ci_entry[ec]=ET_UI; else ci->ci_entry[ec]=ET_I; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TAG_U: if (neg) ci->ci_entry[ec]=ET_UU; else ci->ci_entry[ec]=ET_U; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TAG_CENTER: InsertEntries(ci,align,3); if (neg) { ci->cie_position[ec-2]=18; ci->cie_position[ec-1]=21; } else { ci->cie_position[ec-2]=19; ci->cie_position[ec-1]=21; }; ci->cie_length[ec-2]=1; ci->cie_length[ec-1]=2; break; case TAG_P: ci->ci_entry[ec]=ET_N; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); ci->ci_entry[ec]=ET_N; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TAG_BR: ci->ci_entry[ec]=ET_N; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TAG_OL: case TAG_UL: if (neg) ci->ci_entry[ec]=ET_UUL; else ci->ci_entry[ec]=ET_UL; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; case TAG_LI: ci->ci_entry[ec]=ET_LI; ec++;ecc++;CheckMemory(ci,&ecc,ec,ParserECS); break; }; }; while(*dp && (*dp!='>')) dp++; // skip tag if (*dp=='>') dp++; break; }; }; ci->ci_entry[ec]=ET_END; ci->ci_entrynum=ec; Close(fhan); }; }; return(ci); }