void bbdocument_exist_update_di(char subname[],docid DocID) { // Update DI with new existed timestamp struct DocumentIndexFormat docindex; #if defined(BLACK_BOX) && !defined(_24SEVENOFFICE) DIRead(&docindex, DocID, subname); docindex.lastSeen = time(NULL); DIWrite(&docindex, DocID, subname, NULL); debug("bbdocument_exist: satte lastSeen \"%s\"",ctime(&docindex.lastSeen)); #endif }
int main (int argc, char *argv[]) { struct DocumentIndexFormat DocumentIndexPost; int PopRankextern; int PopRankintern; int PopRanknoc; int PopRanindex; char ShortRank; FILE *FH; struct popl popextern; struct popl popintern; struct popl popnoc; struct popl popindex; uLong htmlBufferSize = 0; char *htmlBuffer = NULL; char *acl_allowbuffer = NULL; char *acl_deniedbuffer = NULL; char timebuf[26]; int optShowhtml = 0; int optShowWords = 0; int optSummary = 0; int optAnchor = 0; int optResource = 0; int optPopRank = 0; int optDelete = 0; int optAdult = 0; unsigned int DocID; char *subname; if (getenv("QUERY_STRING") == NULL) { extern char *optarg; extern int optind, opterr, optopt; char c; while ((c=getopt(argc,argv,"hwsarpdu"))!=-1) { switch (c) { case 'h': optShowhtml = 1; break; case 'u': optAdult = 1; break; case 'w': optShowWords = 1; break; case 's': optSummary = 1; break; case 'a': optAnchor = 1; break; case 'p': optPopRank = 1; break; case 'r': optResource = 1; break; case 'd': optDelete = 1; break; default: exit(1); } } --optind; #ifdef DEBUG printf("argc %i, optind %i\n",argc,optind); #endif if ((argc - optind)!= 3) { printf("Dette programet gir info om en DocID\n\n\tUsage PageInfo DocID collection\n"); exit(1); } DocID = atol(argv[1 +optind]); subname = argv[2 +optind]; } else { printf("Content-type: text/plain\n\n"); int res; // Initialize the CGI lib res = cgi_init(); // Was there an error initializing the CGI??? if (res != CGIERR_NONE) { printf("Error # %d: %s<p>\n", res, cgilib_strerror(res)); fprintf(stderr,"Cgi-lib error."); return -1; } if (cgi_getentrystr("subname") == NULL) { fprintf(stderr,"Didn't recieve any subname."); return -1; } else { subname = cgi_getentrystr("subname"); } if (cgi_getentrystr("DocID") == NULL) { fprintf(stderr,"Didn't recieve any DocID."); return -1; } else { DocID = atol( cgi_getentrystr("DocID") ); } } html_parser_init(); printf("Showing data for Collection \"%s\", DocID %u\n\n",subname,DocID); printf("Lot: %i\n",rLotForDOCid(DocID)); if (optDelete) { memset(&DocumentIndexPost,'\0',sizeof(DocumentIndexPost)); DIWrite(&DocumentIndexPost,DocID,subname,NULL); return 0; } if (DIRead_fmode(&DocumentIndexPost,DocID,subname,'s')) { printf("Url: \"%s\"\nLanguage: %s (id: %s)\nOffensive code: %hu\nDocument type: %s\nTime tested sins last good crawl: %hu\nAdult weight: %hu\nResource size: %u\nIP Address: %u\nHtml size: %i\nImage size: %i\nUser ID: %i\nCrawler version: %f\nRepository pointer: %u\n", DocumentIndexPost.Url, getLangCode2(atoi(DocumentIndexPost.Sprok)), DocumentIndexPost.Sprok, DocumentIndexPost.Offensive_code, DocumentIndexPost.Dokumenttype, DocumentIndexPost.AntallFeiledeCrawl, DocumentIndexPost.AdultWeight, DocumentIndexPost.ResourceSize, DocumentIndexPost.IPAddress, DocumentIndexPost.htmlSize2, DocumentIndexPost.imageSize, DocumentIndexPost.userID, DocumentIndexPost.clientVersion, DocumentIndexPost.RepositoryPointer); if (DocumentIndexPost.response == 200) { printf("HTTP response: %hu\n",DocumentIndexPost.response); } else { printf("HTTP response: \033[1;31m%hu\033[0m\n",DocumentIndexPost.response); } ctime_r((time_t *)&DocumentIndexPost.CrawleDato,timebuf); timebuf[24] = '\0'; printf("Last crawled time: %u\n",DocumentIndexPost.CrawleDato); printf("Last crawled time ISO: %s\n",timebuf); printf("crc32: %u\n",DocumentIndexPost.crc32); #ifdef BLACK_BOX printf("Last seen Unix: %u\n",DocumentIndexPost.lastSeen); printf("Last seen ISO: %s", ctime(&DocumentIndexPost.lastSeen)); #endif printf("Nr of out links: %u\n",(unsigned int)DocumentIndexPost.nrOfOutLinks); char *metadesc, *title, *body; if (DocumentIndexPost.SummarySize == 0) { printf("Summary: Don't have pre-parsed summery (summary size is 0)\n"); } else if (rReadSummary(DocID,&metadesc, &title, &body,DocumentIndexPost.SummaryPointer,DocumentIndexPost.SummarySize,subname)) { printf("\nSummary:\n"); printf("\tSummary pointer: %u\n\tSummary size: %hu\n",DocumentIndexPost.SummaryPointer,DocumentIndexPost.SummarySize); printf("\tTitle from summary: \"%s\"\n\tMeta description from summary: \"%s\"\n",title,metadesc); if (optSummary) { printf("Summary body\n*******************\n%s\n*******************\n\n",body); } } else { printf("Don't have pre-parsed summery\n"); } struct ReposetoryHeaderFormat ReposetoryHeader; char *url, *attributes; if (!rReadHtml(&htmlBuffer,&htmlBufferSize,DocumentIndexPost.RepositoryPointer,DocumentIndexPost.htmlSize2,DocID,subname,&ReposetoryHeader,&acl_allowbuffer,&acl_deniedbuffer,DocumentIndexPost.imageSize, &url, &attributes)) { printf("rReadHtml: did not returne true!\n"); return; } printf("Entire url: %s\n", url); #ifdef BLACK_BOX printf("acl allow raw: \"%s\"\n",acl_allowbuffer); printf("acl denied raw: \"%s\"\n",acl_deniedbuffer); printf("acl allow resolved: \"%s\"\n",aclResolv(acl_allowbuffer)); printf("acl denied resolved: \"%s\"\n",aclResolv(acl_deniedbuffer)); printf("PopRank: %d\n", ReposetoryHeader.PopRank); #endif if (optShowhtml) { printf("html uncompresed size %i\n",htmlBufferSize); printf("html buff:\n*******************************\n"); fwrite(htmlBuffer,htmlBufferSize,1,stdout); printf("\n*******************************\n\n"); } if (optShowWords) { printf("words:\n"); //run_html_parser( DocumentIndexPost.Url, htmlBuffer, htmlBufferSize, fn ); char *title, *body; html_parser_run(url,htmlBuffer, htmlBufferSize,&title, &body,fn,NULL ); } if (optResource) { char buf[500000]; printf("Resource:\n"); printf("Ptr: 0x%x Len: %x\n", DocumentIndexPost.ResourcePointer, DocumentIndexPost.ResourceSize); if (getResource(rLotForDOCid(DocID), subname, DocID, buf, sizeof(buf)) == 0) { printf("\tDid not get any resource\n"); warn(""); } else { printf("%s\n", buf); } } printf("attributes:\"%s\"\n", attributes); free(url); free(attributes); free(acl_allowbuffer); free(acl_deniedbuffer); } else { printf("Cant read post\n"); } #ifndef BLACK_BOX if (optAdult) { int httpResponsCodes[nrOfHttpResponsCodes]; //char *title; //char *body; struct adultFormat *adult; struct pagewordsFormat *pagewords = malloc(sizeof(struct pagewordsFormat)); int AdultWeight; unsigned char langnr; if ((adult = malloc(sizeof(struct adultFormat))) == NULL) { perror("malloc argstruct.adult"); exit(1); } wordsInit(pagewords); langdetectInit(); adultLoad(adult); AdultWeight -1; handelPage(pagewords,&ReposetoryHeader,htmlBuffer,htmlBufferSize,&title,&body); wordsMakeRevIndex(pagewords,adult,&AdultWeight,&langnr); printf("adult %i\n",AdultWeight); } if (optAnchor) { int anchorBufferSize; char *anchorBuffer; anchorBufferSize = anchorRead(rLotForDOCid(DocID),subname,DocID,NULL,-1); anchorBufferSize += 1; anchorBuffer = malloc(anchorBufferSize); anchorRead(rLotForDOCid(DocID),subname,DocID,anchorBuffer,anchorBufferSize); printf("#######################################\nanchors:\n%s\n#######################################\n",anchorBuffer); free(anchorBuffer); } if (optPopRank) { popopen (&popindex,"/home/boitho/config/popindex"); PopRanindex = popRankForDocID(&popindex,DocID); popclose(&popindex); printf("popindex %i\n",PopRanindex); if (popopen (&popextern,"/home/boitho/config/popextern")) { PopRankextern = popRankForDocID(&popextern,DocID); printf("PopRankextern: %i\n",PopRankextern); popclose(&popextern); } if (popopen (&popintern,"/home/boitho/config/popintern")) { PopRankintern = popRankForDocID(&popintern,DocID); printf("PopRankintern %i\n",PopRankintern); popclose(&popintern); } if (popopen (&popnoc,"/home/boitho/config/popnoc")) { PopRanknoc = popRankForDocID(&popnoc,DocID); printf("PopRanknoc %i\n",PopRanknoc); popclose(&popnoc); } if (popopen (&popindex,"/home/boitho/config/popindex")) { PopRanindex = popRankForDocID(&popindex,DocID); printf("popindex %i\n",PopRanindex); popclose(&popindex); } printf("PopRankextern: %i\nPopRankintern %i\nPopRanknoc %i\n",PopRankextern,PopRankintern,PopRanknoc); int brank; popopenMemArray_oneLot(subname,rLotForDOCid(DocID)); brank = popRankForDocIDMemArray(DocID); printf("brank %i\n",brank); //short rank if ( (FH = fopen(SHORTPOPFILE,"rb")) == NULL ) { perror("open"); } else { if ((fseek(FH,DocID* sizeof(ShortRank),SEEK_SET) == 0) && (fread(&ShortRank,sizeof(ShortRank),1,FH) != 0)) { printf("Short rank %u\n",(unsigned char)ShortRank); } else { printf("no hort rank avalibal\n"); }; fclose(FH); } } // if optPopRank #endif }
int main (int argc, char *argv[]) { int lotNr; char lotServer[64]; int pageCount; int i; unsigned int FiltetTime; unsigned int FileOffset; char htmlcompressdbuffer[524288]; //0.5 mb char imagebuffer[524288]; //0.5 mb int httpResponsCodes[nrOfHttpResponsCodes]; struct ReposetoryHeaderFormat ReposetoryHeader; struct DocumentIndexFormat DocumentIndexPost; unsigned long int radress; FILE *revindexFilesHa[NrOfDataDirectorys]; struct adultFormat adult; unsigned int lastIndexTime; if (argc < 3) { printf("Dette programet indekserer en lot. Usage:\n\tIndexerLot lotNr subname\n"); exit(0); } for(i=0;i<nrOfHttpResponsCodes;i++) { httpResponsCodes[i] = 0; } lotNr = atoi(argv[1]); strncpy(subname,argv[2],sizeof(subname) -1); //find server based on lotnr lotlistLoad(); lotlistGetServer(lotServer,lotNr); printf("vil index lot nr %i at %s\n",lotNr,lotServer); adultLoad(&adult); langdetectInit(); //temp: må hente dette fra slot server eller fil FiltetTime = 0; FileOffset = 0; pageCount = 0; if (0) { printf("will ges pages by net\n"); revindexFilesOpenNET(revindexFilesHa); while (rGetNextNET(lotServer,lotNr,&ReposetoryHeader,htmlcompressdbuffer,imagebuffer,&radress,FiltetTime,FileOffset,subname)) { global_curentDocID = ReposetoryHeader.DocID; if (strchr(ReposetoryHeader.url,'?') == 0) { global_curentUrlIsDynamic = 0; } else { global_curentUrlIsDynamic = 1; } handelPage(lotServer,lotNr,&ReposetoryHeader,htmlcompressdbuffer,imagebuffer,revindexFilesHa,&DocumentIndexPost,ReposetoryHeader.DocID,httpResponsCodes,&adult); //datta skal uansett kopieres over //kopierer over di data copyRepToDi(&DocumentIndexPost,&ReposetoryHeader); DocumentIndexPost.RepositoryPointer = radress; //skiver til DocumentIndex DIWriteNET(lotServer,&DocumentIndexPost,ReposetoryHeader.DocID,subname); ++pageCount; //temp: //if(pageCount > 999) { // printf("Exeting after only %i docs\n",pageCount); // break; //} } printf("Sending pages\n"); revindexFilesSendNET(revindexFilesHa,lotNr); } else { printf("Wil acess files localy\n"); //sjekker om vi har nokk palss if (!lotHasSufficientSpace(lotNr,4096,subname)) { printf("insufficient disk space\n"); exit(1); } //finner siste indekseringstid lastIndexTime = GetLastIndexTimeForLot(lotNr,subname); if(lastIndexTime != 0) { printf("lastIndexTime is not 0, but %i\n",lastIndexTime); exit(1); } revindexFilesOpenLocal(revindexFilesHa,lotNr,"Main","wb",subname); //temp:Søker til problemområdet //FileOffset = 334603785; while (rGetNext(lotNr,&ReposetoryHeader,htmlcompressdbuffer,sizeof(htmlcompressdbuffer),imagebuffer,&radress,FiltetTime,FileOffset,subname)) { //printf("D: %u, R: %lu\n",ReposetoryHeader.DocID, radress); global_curentDocID = ReposetoryHeader.DocID; if (strchr(ReposetoryHeader.url,'?') == 0) { global_curentUrlIsDynamic = 0; } else { global_curentUrlIsDynamic = 1; } printf("%s\n",ReposetoryHeader.url); handelPage(lotServer,lotNr,&ReposetoryHeader,htmlcompressdbuffer,imagebuffer,revindexFilesHa,&DocumentIndexPost,ReposetoryHeader.DocID,httpResponsCodes,&adult); //datta skal kopieres over uanset hva som skjer //kopierer over di data copyRepToDi(&DocumentIndexPost,&ReposetoryHeader); DocumentIndexPost.RepositoryPointer = radress; //skiver til DocumentIndex DIWrite(&DocumentIndexPost,ReposetoryHeader.DocID,subname); ++pageCount; //if(pageCount > 9999) { // printf("Exeting after only %i docs\n",pageCount); // //break; // exit(1); //} } //skriver riktig indexstide til lotten setLastIndexTimeForLot(lotNr,httpResponsCodes,subname); // vi må ikke kopiere revindex filene da vi jobber på de lokale direkte } langdetectDestroy(); printf("indexed %i pages\n\n\n",pageCount); return 0; }
void connectHandler(int socket) { struct packedHedderFormat packedHedder; int i,n; int LotNr; char lotPath[512]; char buf[100]; unsigned int FilterTime; int filnamelen; FILE *FH; struct stat inode; // lager en struktur for fstat å returnere. off_t filesize; char c; struct DocumentIndexFormat DocumentIndexPost; int DocID; struct ReposetoryHeaderFormat ReposetoryHeader; unsigned int radress; char htmlbuffer[524288]; int destLeng; char dest[512]; off_t fileBloks,filerest; char *filblocbuff; //while ((i=read(socket, &packedHedder, sizeof(struct packedHedderFormat))) > 0) { while ((i=recv(socket, &packedHedder, sizeof(struct packedHedderFormat),MSG_WAITALL)) > 0) { //printf("command: %i\n",packedHedder.command); //printf("i er %i\n",i); printf("size is: %i\nversion: %i\ncommand: %i\n",packedHedder.size,packedHedder.version,packedHedder.command); //printf("subname: %s\n",packedHedder.subname); //lar size reflektere hva som er igjen av pakken packedHedder.size = packedHedder.size - sizeof(packedHedder); if (packedHedder.command == C_rmkdir) { printf("C_rmkdir\n"); //leser data. Det skal væren en int som sier hvilken lot vi vil ha if ((i=recv(socket, &LotNr, sizeof(LotNr),MSG_WAITALL)) == -1) { perror("Cant read lotnr"); exit(1); } //leser destinasjonelengden if ((i=recv(socket, &destLeng, sizeof(destLeng),MSG_WAITALL)) == -1) { perror("Cant read destLeng"); exit(1); } if (destLeng > sizeof(dest)) { printf("dest filname is to long at %i\n",destLeng); exit(1); } //leser destinasjonene if ((i=recv(socket, &dest, destLeng,MSG_WAITALL)) == -1) { perror("Cant read dest"); exit(1); } GetFilPathForLot(lotPath,LotNr,packedHedder.subname); sprintf(lotPath,"%s%s",lotPath,dest); printf("mkdir %s\n",lotPath); makePath(lotPath); printf("~C_rmkdir\n"); } else if (packedHedder.command == C_rComand) { //leser data. Det skal væren en int som sier hvilken lot vi vil ha if ((i=recv(socket, &LotNr, sizeof(LotNr),MSG_WAITALL)) == -1) { perror("Cant read lotnr"); exit(1); } //leser destinasjonelengden if ((i=recv(socket, &destLeng, sizeof(destLeng),MSG_WAITALL)) == -1) { perror("Cant read destLeng"); exit(1); } if (destLeng > sizeof(dest)) { printf("dest filname is to long at %i\n",destLeng); exit(1); } //leser destinasjonene if ((i=recv(socket, &dest, destLeng,MSG_WAITALL)) == -1) { perror("Cant read dest"); exit(1); } printf("run command %s\n",dest); system(dest); } else if (packedHedder.command == C_getLotToIndex) { printf("fikk C_getLotToIndex\n"); int dirty; if ((i=recv(socket, &dirty, sizeof(dirty),MSG_WAITALL)) == -1) { perror("Cant read dirty"); exit(1); } printf("dirty: %i\n",dirty); LotNr = findLotToIndex(packedHedder.subname,dirty); printf("sending respons\n"); sendall(socket,&LotNr, sizeof(LotNr)); } else if (packedHedder.command == C_getlotHasSufficientSpace) { printf("fikk C_getLotToIndex\n"); int needSpace; int response; if ((i=read(socket, &LotNr, sizeof(LotNr))) == -1) { perror("Cant read lotnr"); exit(1); } if ((i=recv(socket, &needSpace, sizeof(needSpace),MSG_WAITALL)) == -1) { perror("Cant read dirty"); exit(1); } printf("needSpace: %i, LotNr %i\n",needSpace,LotNr); response = lotHasSufficientSpace(LotNr, needSpace, packedHedder.subname); printf("sending respons\n"); sendall(socket,&response, sizeof(response)); } else if (packedHedder.command == C_rGetSize) { printf("fikk C_rGetSize\n"); //leser data. Det skal væren en int som sier hvilken lot vi vil ha if ((i=read(socket, &LotNr, sizeof(LotNr))) == -1) { perror("Cant read lotnr"); exit(1); } if ((i=read(socket, &filnamelen, sizeof(filnamelen))) == -1) { perror("Cant read filnamelen"); exit(1); } if (filnamelen > sizeof(buf)) { printf("filname to long\n"); }; if ((i=read(socket, buf, filnamelen)) == -1) { perror("Cant read filnamelen"); exit(1); } printf("filname %s\n",buf); if ((FH = lotOpenFileNoCasheByLotNr(LotNr,buf,"rb",'s',packedHedder.subname)) == NULL) { perror(buf); //sending that he fil is emty fileBloks = 0; sendall(socket,&fileBloks, sizeof(fileBloks)); } else { //finner og sender il størelse fstat(fileno(FH),&inode); //filesize = inode.st_size; //sendall(socket,&filesize, sizeof(filesize)); fileBloks = inode.st_size; printf("size is %" PRId64 "\n",fileBloks); sendall(socket,&fileBloks, sizeof(fileBloks)); fclose(FH); } } else if (packedHedder.command == C_rGetFile) { printf("fikk C_rGetFile\n"); //leser data. Det skal væren en int som sier hvilken lot vi vil ha if ((i=read(socket, &LotNr, sizeof(LotNr))) == -1) { perror("Cant read lotnr"); exit(1); } if ((i=read(socket, &filnamelen, sizeof(filnamelen))) == -1) { perror("Cant read filnamelen"); exit(1); } if (filnamelen > sizeof(buf)) { printf("filname to long\n"); }; if ((i=read(socket, buf, filnamelen)) == -1) { perror("Cant read filnamelen"); exit(1); } printf("filname %s\n",buf); if ((FH = lotOpenFileNoCasheByLotNr(LotNr,buf,"rb",'s',packedHedder.subname)) == NULL) { perror(buf); //sending that the fil is emty fileBloks = 0; filerest = 0; sendall(socket,&fileBloks, sizeof(fileBloks)); sendall(socket,&filerest, sizeof(filerest)); } else { //finner og sender fil størelse fstat(fileno(FH),&inode); //filesize = inode.st_size; //sendall(socket,&filesize, sizeof(filesize)); fileBloks = (int)floor(inode.st_size / rNetTrabsferBlok); filerest = inode.st_size - (fileBloks * rNetTrabsferBlok); sendall(socket,&fileBloks, sizeof(fileBloks)); sendall(socket,&filerest, sizeof(filerest)); printf("sending fil. fileBloks %"PRId64", filerest %"PRId64"\n",fileBloks,filerest); filblocbuff = (char *)malloc(rNetTrabsferBlok); for(i=0; i < fileBloks; i++) { //fread(filblocbuff,sizeof(c),rNetTrabsferBlok,FH); //fread_all(const void *buf, size_t size, FILE *stream) fread_all(filblocbuff,rNetTrabsferBlok,FH, 4096); if ((n=sendall(socket, filblocbuff, rNetTrabsferBlok)) == -1) { perror("Cant recv dest"); exit(1); } } printf("did recv %i fileBloks\n",i); fread(filblocbuff,sizeof(c),filerest,FH); if ((n=sendall(socket, filblocbuff, filerest)) == -1) { perror("Cant recv filerest"); exit(1); } free(filblocbuff); /* for (i=0;i<filesize;i++) { fread(&c,sizeof(char),1,FH); send(socket, &c, sizeof(char), 0); //printf("%i\n",(int)c); } */ printf("send file end\n"); fclose(FH); } } else if (packedHedder.command == C_rGetNext) { printf("fikk C_rGetNext\n"); printf("støttes ikke lengere"); exit(1); /* //leser data. Det skal væren en unigned int som sier hvilken lot vi vil ha //har deklarert den som int her ??? if ((i=read(socket, &LotNr, sizeof(LotNr))) == -1) { perror("Cant read lotnr"); exit(1); } printf("leser FilterTime\n"); //leser filtertime if ((i=read(socket, &FilterTime, sizeof(FilterTime))) == -1) { perror("Cant read lotnr"); exit(1); } printf("lotnr %i FilterTime %u\n",LotNr,FilterTime); //henter inn data om den lotten if (rGetNext(LotNr,&ReposetoryHeader,htmlbuffer,NULL,&radress,FilterTime,0)) { //printf("DocId: %i url: %s\n",ReposetoryHeader.DocID,ReposetoryHeader.url); //sender pakke hedder sendpacked(socket,C_rLotData,PROTOCOLVERSION, ReposetoryHeader.htmlSize + sizeof(ReposetoryHeader) +sizeof(radress), NULL,packedHedder.subname); //sennder ReposetoryHeader'en sendall(socket,&ReposetoryHeader, sizeof(ReposetoryHeader)); //sender htmlen sendall(socket,&htmlbuffer, ReposetoryHeader.htmlSize); //sender adressen sendall(socket,&radress,sizeof(radress)); //printf("data sent\n"); //printf("rGetNext: %i\n",ReposetoryHeader.DocID); } else { sendpacked(socket,C_rEOF,PROTOCOLVERSION, 0, NULL,packedHedder.subname); printf("ferdig\n"); } */ } else if (packedHedder.command == C_DIWrite) { if ((i=recv(socket, &DocumentIndexPost, sizeof(struct DocumentIndexFormat),MSG_WAITALL)) == -1) { perror("recv"); exit(1); } if ((i=recv(socket, &DocID, sizeof(DocID),MSG_WAITALL)) == -1) { perror("recv"); exit(1); } DIWrite(&DocumentIndexPost,DocID,packedHedder.subname, NULL); //printf("DIWrite: %i\n",DocID); } else if (packedHedder.command == C_DIRead) { int DocID; struct DocumentIndexFormat DocumentIndexPost; printf("got commane C_DIRead. sise %i hsize %i ds %i\n",packedHedder.size, sizeof(packedHedder), sizeof(DocID)); if ((i=recv(socket, &DocID, sizeof(DocID),0)) == -1) { perror("recv"); exit(1); } //printf("DocID %i\n",DocID); //leser inn datan //int DIRead (struct DocumentIndexFormat *DocumentIndexPost, int DocID); DIRead(&DocumentIndexPost,DocID,packedHedder.subname); sendall(socket,&DocumentIndexPost, sizeof(struct DocumentIndexFormat)); } else if (packedHedder.command == C_rGetIndexTime) { int Lotnr; unsigned int IndexTime; if ((i=recv(socket, &LotNr, sizeof(LotNr),0)) == -1) { perror("recv"); exit(1); } IndexTime = GetLastIndexTimeForLot(LotNr,packedHedder.subname); sendall(socket,&IndexTime, sizeof(IndexTime)); } else if (packedHedder.command == C_rSetIndexTime) { int Lotnr; if ((i=recv(socket, &LotNr, sizeof(LotNr),0)) == -1) { perror("recv"); exit(1); } setLastIndexTimeForLot(LotNr,NULL,packedHedder.subname); } else if (packedHedder.command == C_rSendFile) { //skal mota en fil for lagring i reposetoryet //char FilePath[156]; FILE *FILEHANDLER; char c; char opentype[2]; //char *filblocbuff; //off_t fileBloks,filerest; if ((i=recv(socket, &LotNr, sizeof(LotNr),MSG_WAITALL)) == -1) { perror("Cant recv lotnr"); exit(1); } printf("lotNr %i\n",LotNr); //leser destinasjonelengden if ((i=recv(socket, &destLeng, sizeof(destLeng),MSG_WAITALL)) == -1) { perror("Cant recv destLeng"); exit(1); } if (destLeng > sizeof(dest)) { printf("dest filname is to long at %i\n",destLeng); exit(1); } //leser destinasjonene if ((i=recv(socket, &dest, destLeng,MSG_WAITALL)) == -1) { perror("Cant recv dest"); exit(1); } printf("coping %s as length %i in to lot %i\n",dest,destLeng,LotNr); if ((i=recv(socket, &opentype, sizeof(char) +1,MSG_WAITALL)) == -1) { perror("Cant recv opentype"); exit(1); } printf("opentype \"%s\"\n",opentype); //GetFilPathForLot(FilePath,LotNr,packedHedder.subname); //legger til filnavnet //strncat(FilePath,dest,sizeof(FilePath)); //leser inn filstørelsen if ((i=recv(socket, &fileBloks, sizeof(fileBloks),MSG_WAITALL)) == -1) { perror("Cant recv fileBloks"); exit(1); } if ((i=recv(socket, &filerest, sizeof(filerest),MSG_WAITALL)) == -1) { perror("Cant recv filerest"); exit(1); } printf("fileBloks: %" PRId64 ", filerest: %" PRId64 "\n",fileBloks,filerest); //åpner filen if ((FILEHANDLER = lotOpenFileNoCasheByLotNr(LotNr,dest,opentype,'e',packedHedder.subname)) == NULL) { perror(dest); } filblocbuff = (char *)malloc(rNetTrabsferBlok); for(i=0; i < fileBloks; i++) { if ((n=recv(socket, filblocbuff, rNetTrabsferBlok,MSG_WAITALL)) == -1) { perror("Cant recv dest"); exit(1); } fwrite(filblocbuff,sizeof(c),rNetTrabsferBlok,FILEHANDLER); } printf("did recv %i fileBloks\n",i); if ((n=recv(socket, filblocbuff, filerest,MSG_WAITALL)) == -1) { perror("Cant recv filerest"); exit(1); } fwrite(filblocbuff,sizeof(c),filerest,FILEHANDLER); free(filblocbuff); fclose(FILEHANDLER); printf("\n"); } else if (packedHedder.command == C_DIGetIp) { unsigned int DocID; struct DocumentIndexFormat DocumentIndexPost; //printf("got command C_DIGetIp\n"); if ((i=recv(socket, &DocID, sizeof(DocID),MSG_WAITALL)) == -1) { perror("recv"); exit(1); } //printf("DocID %u\n",DocID); DIRead(&DocumentIndexPost,DocID,packedHedder.subname); //printf("ipadress: %u\n",DocumentIndexPost.IPAddress); sendall(socket,&DocumentIndexPost.IPAddress, sizeof(DocumentIndexPost.IPAddress)); } else if (packedHedder.command == C_anchorAdd) { size_t textlen; unsigned int DocID; char *text; printf("Add anchor....\n"); if ((i = recv(socket, &DocID, sizeof(DocID),MSG_WAITALL)) == -1) { perror("recv"); exit(1); } else if ((i = recv(socket, &textlen, sizeof(textlen), MSG_WAITALL)) == -1) { perror("recv(textlen)"); exit(1); } text = malloc(textlen+1); text[textlen] = '\0'; if ((i = recv(socket, text, textlen, MSG_WAITALL)) == -1) { perror("recv(text)"); exit(1); } anchoraddnew(DocID, text, textlen, packedHedder.subname, NULL); printf("Text for %d: %s\n", DocID, text); free(text); } else if (packedHedder.command == C_anchorGet) { size_t len; char *text; int LotNr; unsigned int DocID; printf("Get anchor...\n"); if ((i = recv(socket, &DocID, sizeof(DocID),MSG_WAITALL)) == -1) { perror("recv"); exit(1); } printf("got DocID %u\n",DocID); LotNr = rLotForDOCid(DocID); printf("trying to read anchor\n"); len = anchorRead(LotNr, packedHedder.subname, DocID, NULL, -1); printf("got anchor of length %i\n",len); sendall(socket, &len, sizeof(len)); text = malloc(len+1); printf("readint it again\n"); anchorRead(LotNr, packedHedder.subname, DocID, text, len+1); sendall(socket, text, len); } else if (packedHedder.command == C_readHTML) { /* unsigned int DocID; unsigned int len; char *text; char *acla, *acld; struct DocumentIndexFormat DocIndex; struct ReposetoryHeaderFormat ReposetoryHeader; if ((i = recv(socket, &DocID, sizeof(DocID), MSG_WAITALL)) == -1) { perror("recv"); exit(1); } if ((i = recv(socket, &len, sizeof(len), MSG_WAITALL)) == -1) { perror("recv(len)"); exit(1); } printf("len %u\n",len); text = malloc(len); if (text == NULL) exit(1); DIRead(&DocIndex, DocID, packedHedder.subname); if (!rReadHtml( text, &len, DocIndex.RepositoryPointer, DocIndex.htmlSize, DocID, packedHedder.subname, &ReposetoryHeader, &acla, &acld, DocIndex.imageSize)) { len = 0; sendall(socket, &len, sizeof(len)); } else { ++len; // \0 #ifdef DEBUG printf("docID %u\n",DocID); printf("Got: (len %i, real %i) ########################\n%s\n#####################\n", len, strlen(text), text); #endif sendall(socket, &len, sizeof(len)); sendall(socket, text, len); sendall(socket, &ReposetoryHeader,sizeof(ReposetoryHeader)); } free(text); */ } /* runarb: 06 des 2007: vi har gåt bort fra denne metoden for nå, og bruker heller index over smb. Men tar vare på den da vi kan trenge den siden else if (packedHedder.command == C_urltodocid) { char cmd; int alloclen; char *urlbuf; if (urltodociddb == NULL) { cmd = C_DOCID_NODB; sendall(socket, &cmd, sizeof(cmd)); exit(1); } else { cmd = C_DOCID_READY; sendall(socket, &cmd, sizeof(cmd)); } cmd = C_DOCID_NEXT; alloclen = 1024; urlbuf = malloc(alloclen); do { unsigned int DocID; size_t len; if ((i = recv(socket, &cmd, sizeof(cmd), MSG_WAITALL)) == -1) { err(1, "recv(cmd)"); } if (cmd == C_DOCID_DONE) break; if ((i == recv(socket, &len, sizeof(len), MSG_WAITALL)) == -1) { err(1, "recv(len)"); } if (alloclen < len+1) { free(urlbuf); alloclen *= 2; urlbuf = malloc(alloclen); } if ((i == recv(socket, urlbuf, len, MSG_WAITALL)) == -1) { err(1, "recv(len)"); } urlbuf[len] = '\0'; if (!getDocIDFromUrl(urltodociddb, urlbuf, &DocID)) { cmd = C_DOCID_NOTFOUND; sendall(socket, &cmd, sizeof(cmd)); } else { cmd = C_DOCID_FOUND; sendall(socket, &cmd, sizeof(cmd)); sendall(socket, &DocID, sizeof(DocID)); } } while (1); free(urlbuf); } */ else { printf("unnown comand. %i\n", packedHedder.command); } //printf("size is: %i\nversion: %i\ncommand: %i\n",packedHedder.size,packedHedder.version,packedHedder.command); } //while }
int main (int argc, char *argv[]) { int lotNr; char lotServer[64]; int pageCount; int i; unsigned int FiltetTime; unsigned int FileOffset; char htmlcompressdbuffer[524288]; //0.5 mb char imagebuffer[524288]; //0.5 mb int httpResponsCodes[nrOfHttpResponsCodes]; struct ReposetoryHeaderFormat ReposetoryHeader; struct DocumentIndexFormat DocumentIndexPost; unsigned long int radress; FILE *revindexFilesHa[NrOfDataDirectorys]; struct adultFormat adult; unsigned int lastIndexTime; if (argc < 2) { printf("Dette programet indekserer en lot. Gi det et lot nummer\n"); exit(0); } for(i=0;i<nrOfHttpResponsCodes;i++) { httpResponsCodes[i] = 0; } lotNr = atoi(argv[1]); //find server based on lotnr lotlistLoad(); lotlistGetServer(lotServer,lotNr); printf("vil index lot nr %i at %s\n",lotNr,lotServer); //finner siste indekseringstid lastIndexTime = GetLastIndexTimeForLot(lotNr,subname); if(lastIndexTime == 0) { printf("lastIndexTime is 0, skiping.\n"); exit(1); } //temp: må hente dette fra slot server eller fil FiltetTime = 0; FileOffset = 2140483648; //FileOffset = 1997015914; pageCount = 0; printf("Wil acess files localy\n"); while (rGetNext(lotNr,&ReposetoryHeader,htmlcompressdbuffer,imagebuffer,&radress,FiltetTime,FileOffset,subname)) { DIRead(&DocumentIndexPost,ReposetoryHeader.DocID,subname); DocumentIndexPost.RepositoryPointer = radress; //skiver til DocumentIndex DIWrite(&DocumentIndexPost,ReposetoryHeader.DocID,subname); ++pageCount; } printf("indexed %i pages\n\n\n",pageCount); return 0; }