/* Copyright 2012 Mozilla Foundation * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ import { createPromiseCapability } from './pdfjs'; import { scrollIntoView } from './ui_utils'; var FindStates = { FIND_FOUND: 0, FIND_NOTFOUND: 1, FIND_WRAPPED: 2, FIND_PENDING: 3 }; var FIND_SCROLL_OFFSET_TOP = -50; var FIND_SCROLL_OFFSET_LEFT = -400; var CHARACTERS_TO_NORMALIZE = { '\u2018': '\'', // Left single quotation mark '\u2019': '\'', // Right single quotation mark '\u201A': '\'', // Single low-9 quotation mark '\u201B': '\'', // Single high-reversed-9 quotation mark '\u201C': '"', // Left double quotation mark '\u201D': '"', // Right double quotation mark '\u201E': '"', // Double low-9 quotation mark '\u201F': '"', // Double high-reversed-9 quotation mark '\u00BC': '1/4', // Vulgar fraction one quarter '\u00BD': '1/2', // Vulgar fraction one half '\u00BE': '3/4', // Vulgar fraction three quarters }; /** * Provides "search" or "find" functionality for the PDF. * This object actually performs the search for a given string. */ var PDFFindController = (function PDFFindControllerClosure() { function PDFFindController(options) { this.pdfViewer = options.pdfViewer || null; this.onUpdateResultsCount = null; this.onUpdateState = null; this.reset(); // Compile the regular expression for text normalization once. var replace = Object.keys(CHARACTERS_TO_NORMALIZE).join(''); this.normalizationRegex = new RegExp('[' + replace + ']', 'g'); } PDFFindController.prototype = { reset: function PDFFindController_reset() { this.startedTextExtraction = false; this.extractTextPromises = []; this.pendingFindMatches = Object.create(null); this.active = false; // If active, find results will be highlighted. this.pageContents = []; // Stores the text for each page. this.pageMatches = []; this.pageMatchesLength = null; this.matchCount = 0; this.selected = { // Currently selected match. pageIdx: -1, matchIdx: -1 }; this.offset = { // Where the find algorithm currently is in the document. pageIdx: null, matchIdx: null }; this.pagesToSearch = null; this.resumePageIdx = null; this.state = null; this.dirtyMatch = false; this.findTimeout = null; this._firstPagePromise = new Promise((resolve) => { this.resolveFirstPage = resolve; }); }, normalize: function PDFFindController_normalize(text) { return text.replace(this.normalizationRegex, function (ch) { return CHARACTERS_TO_NORMALIZE[ch]; }); }, // Helper for multiple search - fills matchesWithLength array // and takes into account cases when one search term // include another search term (for example, "tamed tame" or "this is"). // Looking for intersecting terms in the 'matches' and // leave elements with a longer match-length. _prepareMatches: function PDFFindController_prepareMatches( matchesWithLength, matches, matchesLength) { function isSubTerm(matchesWithLength, currentIndex) { var currentElem, prevElem, nextElem; currentElem = matchesWithLength[currentIndex]; nextElem = matchesWithLength[currentIndex + 1]; // checking for cases like "TAMEd TAME" if (currentIndex < matchesWithLength.length - 1 && currentElem.match === nextElem.match) { currentElem.skipped = true; return true; } // checking for cases like "thIS IS" for (var i = currentIndex - 1; i >= 0; i--) { prevElem = matchesWithLength[i]; if (prevElem.skipped) { continue; } if (prevElem.match + prevElem.matchLength < currentElem.match) { break; } if (prevElem.match + prevElem.matchLength >= currentElem.match + currentElem.matchLength) { currentElem.skipped = true; return true; } } return false; } var i, len; // Sorting array of objects { match: , matchLength: } // in increasing index first and then the lengths. matchesWithLength.sort(function(a, b) { return a.match === b.match ? a.matchLength - b.matchLength : a.match - b.match; }); for (i = 0, len = matchesWithLength.length; i < len; i++) { if (isSubTerm(matchesWithLength, i)) { continue; } matches.push(matchesWithLength[i].match); matchesLength.push(matchesWithLength[i].matchLength); } }, calcFindPhraseMatch: function PDFFindController_calcFindPhraseMatch( query, pageIndex, pageContent) { var matches = []; var queryLen = query.length; var matchIdx = -queryLen; while (true) { matchIdx = pageContent.indexOf(query, matchIdx + queryLen); if (matchIdx === -1) { break; } matches.push(matchIdx); } this.pageMatches[pageIndex] = matches; }, calcFindWordMatch: function PDFFindController_calcFindWordMatch( query, pageIndex, pageContent) { var matchesWithLength = []; // Divide the query into pieces and search for text on each piece. var queryArray = query.match(/\S+/g); var subquery, subqueryLen, matchIdx; for (var i = 0, len = queryArray.length; i < len; i++) { subquery = queryArray[i]; subqueryLen = subquery.length; matchIdx = -subqueryLen; while (true) { matchIdx = pageContent.indexOf(subquery, matchIdx + subqueryLen); if (matchIdx === -1) { break; } // Other searches do not, so we store the length. matchesWithLength.push({ match: matchIdx, matchLength: subqueryLen, skipped: false }); } } // Prepare arrays for store the matches. if (!this.pageMatchesLength) { this.pageMatchesLength = []; } this.pageMatchesLength[pageIndex] = []; this.pageMatches[pageIndex] = []; // Sort matchesWithLength, clean up intersecting terms // and put the result into the two arrays. this._prepareMatches(matchesWithLength, this.pageMatches[pageIndex], this.pageMatchesLength[pageIndex]); }, calcFindMatch: function PDFFindController_calcFindMatch(pageIndex) { var pageContent = this.normalize(this.pageContents[pageIndex]); var query = this.normalize(this.state.query); var caseSensitive = this.state.caseSensitive; var phraseSearch = this.state.phraseSearch; var queryLen = query.length; if (queryLen === 0) { // Do nothing: the matches should be wiped out already. return; } if (!caseSensitive) { pageContent = pageContent.toLowerCase(); query = query.toLowerCase(); } if (phraseSearch) { this.calcFindPhraseMatch(query, pageIndex, pageContent); } else { this.calcFindWordMatch(query, pageIndex, pageContent); } this.updatePage(pageIndex); if (this.resumePageIdx === pageIndex) { this.resumePageIdx = null; this.nextPageMatch(); } // Update the matches count if (this.pageMatches[pageIndex].length > 0) { this.matchCount += this.pageMatches[pageIndex].length; this.updateUIResultsCount(); } }, extractText() { if (this.startedTextExtraction) { return; } this.startedTextExtraction = true; this.pageContents.length = 0; let promise = Promise.resolve(); for (let i = 0, ii = this.pdfViewer.pagesCount; i < ii; i++) { let extractTextCapability = createPromiseCapability(); this.extractTextPromises[i] = extractTextCapability.promise; promise = promise.then(() => { return this.pdfViewer.getPageTextContent(i).then((textContent) => { let textItems = textContent.items; let strBuf = []; for (let j = 0, jj = textItems.length; j < jj; j++) { strBuf.push(textItems[j].str); } // Store the pageContent as a string. this.pageContents[i] = strBuf.join(''); extractTextCapability.resolve(i); }); }); } }, executeCommand: function PDFFindController_executeCommand(cmd, state) { if (this.state === null || cmd !== 'findagain') { this.dirtyMatch = true; } this.state = state; this.updateUIState(FindStates.FIND_PENDING); this._firstPagePromise.then(() => { this.extractText(); clearTimeout(this.findTimeout); if (cmd === 'find') { // Only trigger the find action after 250ms of silence. this.findTimeout = setTimeout(this.nextMatch.bind(this), 250); } else { this.nextMatch(); } }); }, updatePage: function PDFFindController_updatePage(index) { if (this.selected.pageIdx === index) { // If the page is selected, scroll the page into view, which triggers // rendering the page, which adds the textLayer. Once the textLayer is // build, it will scroll onto the selected match. this.pdfViewer.currentPageNumber = index + 1; } var page = this.pdfViewer.getPageView(index); if (page.textLayer) { page.textLayer.updateMatches(); } }, nextMatch: function PDFFindController_nextMatch() { var previous = this.state.findPrevious; var currentPageIndex = this.pdfViewer.currentPageNumber - 1; var numPages = this.pdfViewer.pagesCount; this.active = true; if (this.dirtyMatch) { // Need to recalculate the matches, reset everything. this.dirtyMatch = false; this.selected.pageIdx = this.selected.matchIdx = -1; this.offset.pageIdx = currentPageIndex; this.offset.matchIdx = null; this.hadMatch = false; this.resumePageIdx = null; this.pageMatches = []; this.matchCount = 0; this.pageMatchesLength = null; for (let i = 0; i < numPages; i++) { // Wipe out any previous highlighted matches. this.updatePage(i); // As soon as the text is extracted start finding the matches. if (!(i in this.pendingFindMatches)) { this.pendingFindMatches[i] = true; this.extractTextPromises[i].then((pageIdx) => { delete this.pendingFindMatches[pageIdx]; this.calcFindMatch(pageIdx); }); } } } // If there's no query there's no point in searching. if (this.state.query === '') { this.updateUIState(FindStates.FIND_FOUND); return; } // If we're waiting on a page, we return since we can't do anything else. if (this.resumePageIdx) { return; } var offset = this.offset; // Keep track of how many pages we should maximally iterate through. this.pagesToSearch = numPages; // If there's already a matchIdx that means we are iterating through a // page's matches. if (offset.matchIdx !== null) { var numPageMatches = this.pageMatches[offset.pageIdx].length; if ((!previous && offset.matchIdx + 1 < numPageMatches) || (previous && offset.matchIdx > 0)) { // The simple case; we just have advance the matchIdx to select // the next match on the page. this.hadMatch = true; offset.matchIdx = (previous ? offset.matchIdx - 1 : offset.matchIdx + 1); this.updateMatch(true); return; } // We went beyond the current page's matches, so we advance to // the next page. this.advanceOffsetPage(previous); } // Start searching through the page. this.nextPageMatch(); }, matchesReady: function PDFFindController_matchesReady(matches) { var offset = this.offset; var numMatches = matches.length; var previous = this.state.findPrevious; if (numMatches) { // There were matches for the page, so initialize the matchIdx. this.hadMatch = true; offset.matchIdx = (previous ? numMatches - 1 : 0); this.updateMatch(true); return true; } // No matches, so attempt to search the next page. this.advanceOffsetPage(previous); if (offset.wrapped) { offset.matchIdx = null; if (this.pagesToSearch < 0) { // No point in wrapping again, there were no matches. this.updateMatch(false); // while matches were not found, searching for a page // with matches should nevertheless halt. return true; } } // Matches were not found (and searching is not done). return false; }, /** * The method is called back from the text layer when match presentation * is updated. * @param {number} pageIndex - page index. * @param {number} index - match index. * @param {Array} elements - text layer div elements array. * @param {number} beginIdx - start index of the div array for the match. */ updateMatchPosition: function PDFFindController_updateMatchPosition( pageIndex, index, elements, beginIdx) { if (this.selected.matchIdx === index && this.selected.pageIdx === pageIndex) { var spot = { top: FIND_SCROLL_OFFSET_TOP, left: FIND_SCROLL_OFFSET_LEFT }; scrollIntoView(elements[beginIdx], spot, /* skipOverflowHiddenElements = */ true); } }, nextPageMatch: function PDFFindController_nextPageMatch() { if (this.resumePageIdx !== null) { console.error('There can only be one pending page.'); } do { var pageIdx = this.offset.pageIdx; var matches = this.pageMatches[pageIdx]; if (!matches) { // The matches don't exist yet for processing by "matchesReady", // so set a resume point for when they do exist. this.resumePageIdx = pageIdx; break; } } while (!this.matchesReady(matches)); }, advanceOffsetPage: function PDFFindController_advanceOffsetPage(previous) { var offset = this.offset; var numPages = this.extractTextPromises.length; offset.pageIdx = (previous ? offset.pageIdx - 1 : offset.pageIdx + 1); offset.matchIdx = null; this.pagesToSearch--; if (offset.pageIdx >= numPages || offset.pageIdx < 0) { offset.pageIdx = (previous ? numPages - 1 : 0); offset.wrapped = true; } }, updateMatch: function PDFFindController_updateMatch(found) { var state = FindStates.FIND_NOTFOUND; var wrapped = this.offset.wrapped; this.offset.wrapped = false; if (found) { var previousPage = this.selected.pageIdx; this.selected.pageIdx = this.offset.pageIdx; this.selected.matchIdx = this.offset.matchIdx; state = (wrapped ? FindStates.FIND_WRAPPED : FindStates.FIND_FOUND); // Update the currently selected page to wipe out any selected matches. if (previousPage !== -1 && previousPage !== this.selected.pageIdx) { this.updatePage(previousPage); } } this.updateUIState(state, this.state.findPrevious); if (this.selected.pageIdx !== -1) { this.updatePage(this.selected.pageIdx); } }, updateUIResultsCount: function PDFFindController_updateUIResultsCount() { if (this.onUpdateResultsCount) { this.onUpdateResultsCount(this.matchCount); } }, updateUIState: function PDFFindController_updateUIState(state, previous) { if (this.onUpdateState) { this.onUpdateState(state, previous, this.matchCount); } } }; return PDFFindController; })(); export { FindStates, PDFFindController, };