[855] | 1 | /**************************************************************************
|
---|
| 2 | *
|
---|
| 3 | * TextEl.cpp -- Data structures for parsed documents
|
---|
| 4 | * Copyright (C) 1999 Rodger McNab
|
---|
| 5 | *
|
---|
| 6 | * This program is free software; you can redistribute it and/or modify
|
---|
| 7 | * it under the terms of the GNU General Public License as published by
|
---|
| 8 | * the Free Software Foundation; either version 2 of the License, or
|
---|
| 9 | * (at your option) any later version.
|
---|
| 10 | *
|
---|
| 11 | * This program is distributed in the hope that it will be useful,
|
---|
| 12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 14 | * GNU General Public License for more details.
|
---|
| 15 | *
|
---|
| 16 | * You should have received a copy of the GNU General Public License
|
---|
| 17 | * along with this program; if not, write to the Free Software
|
---|
| 18 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 19 | *
|
---|
| 20 | * $Id: TextEl.cpp 855 2000-01-14 02:17:52Z sjboddie $
|
---|
| 21 | *
|
---|
| 22 | **************************************************************************/
|
---|
| 23 |
|
---|
| 24 | #include "TextEl.h"
|
---|
| 25 | #include "ctype.h"
|
---|
| 26 |
|
---|
| 27 | #define REC_TERM '\002'
|
---|
| 28 | #define PARA_TERM '\003'
|
---|
| 29 |
|
---|
| 30 |
|
---|
| 31 | void TextEl::Clear () {
|
---|
| 32 | elType = TextE;
|
---|
| 33 | tagName.erase (tagName.begin(), tagName.end());
|
---|
| 34 | text.erase (text.begin(), text.end());
|
---|
| 35 | }
|
---|
| 36 |
|
---|
| 37 |
|
---|
| 38 | static bool ReadTextTag (FileBuf &buf, TextEl &el) {
|
---|
| 39 | // we must have some sort of tag
|
---|
| 40 | unsigned char c;
|
---|
| 41 | if (!buf.Peek(c)) return false;
|
---|
| 42 |
|
---|
| 43 | // get the '<'
|
---|
| 44 | if (c != '<') return false;
|
---|
| 45 | el.text.push_back (c);
|
---|
| 46 | if (!buf.Next() || !buf.Peek (c)) return false;
|
---|
| 47 |
|
---|
| 48 | // see if we have a '/'
|
---|
| 49 | if (c == '/') {
|
---|
| 50 | // we have a closing tag
|
---|
| 51 | el.text.push_back (c);
|
---|
| 52 | if (!buf.Next() || !buf.Peek (c)) return false;
|
---|
| 53 | el.elType = CloseTagE;
|
---|
| 54 | } else {
|
---|
| 55 | // we have an opening tag
|
---|
| 56 | el.elType = OpenTagE;
|
---|
| 57 | }
|
---|
| 58 |
|
---|
| 59 | // get the tag name
|
---|
| 60 | while (isalnum(c)) {
|
---|
| 61 | el.text.push_back (c);
|
---|
| 62 | el.tagName.push_back (c);
|
---|
| 63 | if (!buf.Next() || !buf.Peek (c)) return false;
|
---|
| 64 | }
|
---|
| 65 |
|
---|
| 66 | // get everything up to the closing '>'
|
---|
| 67 | while (c != '>') {
|
---|
| 68 | el.text.push_back (c);
|
---|
| 69 | if (!buf.Next() || !buf.Peek (c)) return false;
|
---|
| 70 | }
|
---|
| 71 |
|
---|
| 72 | // get the closing '>'
|
---|
| 73 | el.text.push_back (c);
|
---|
| 74 | buf.Next();
|
---|
| 75 |
|
---|
| 76 | // if there is no tag name then make this a text element
|
---|
| 77 | if (el.tagName.empty()) el.elType = TextE;
|
---|
| 78 |
|
---|
| 79 | return true;
|
---|
| 80 | }
|
---|
| 81 |
|
---|
| 82 | static void ToggleParaTag (TextEl &el, bool &compatInPara) {
|
---|
| 83 | SetCStr (el.tagName, "Paragraph");
|
---|
| 84 | el.text.erase (el.text.begin(), el.text.end());
|
---|
| 85 | if (compatInPara) {
|
---|
| 86 | el.elType = CloseTagE;
|
---|
| 87 | el.text.push_back (PARA_TERM);
|
---|
| 88 | } else el.elType = OpenTagE;
|
---|
| 89 | compatInPara = !compatInPara;
|
---|
| 90 | }
|
---|
| 91 |
|
---|
| 92 | static void SetRecTag (TextEl &el, TextElType elType) {
|
---|
| 93 | el.elType = elType;
|
---|
| 94 | SetCStr (el.tagName, "Document");
|
---|
| 95 | el.text.erase (el.text.begin(), el.text.end());
|
---|
| 96 | if (elType == CloseTagE)
|
---|
| 97 | el.text.push_back (REC_TERM);
|
---|
| 98 | }
|
---|
| 99 |
|
---|
| 100 |
|
---|
| 101 | bool ReadTextEl (FileBuf &buf, TextEl &el,
|
---|
| 102 | bool compatMode, bool &compatInPara) {
|
---|
| 103 | el.Clear();
|
---|
| 104 |
|
---|
| 105 | unsigned char c;
|
---|
| 106 | if (!buf.Peek (c)) return false;
|
---|
| 107 |
|
---|
| 108 | if (compatMode) {
|
---|
| 109 | if (c == PARA_TERM) {
|
---|
| 110 | ToggleParaTag (el, compatInPara);
|
---|
| 111 | if (!compatInPara) buf.Next();
|
---|
| 112 | return true;
|
---|
| 113 | }
|
---|
| 114 | if (c == REC_TERM) {
|
---|
| 115 | if (compatInPara) {
|
---|
| 116 | // need to close this paragraph
|
---|
| 117 | ToggleParaTag (el, compatInPara);
|
---|
| 118 | return true;
|
---|
| 119 | }
|
---|
| 120 | // can close this document
|
---|
| 121 | buf.Next();
|
---|
| 122 | SetRecTag (el, CloseTagE);
|
---|
| 123 | return true;
|
---|
| 124 | }
|
---|
| 125 | if (!compatInPara) {
|
---|
| 126 | // need to open a paragraph
|
---|
| 127 | ToggleParaTag (el, compatInPara);
|
---|
| 128 | return true;
|
---|
| 129 | }
|
---|
| 130 | }
|
---|
| 131 |
|
---|
| 132 | // see if we have some sort of tag
|
---|
| 133 | if (c == '<') return ReadTextTag (buf, el);
|
---|
| 134 |
|
---|
| 135 | // read in a text element
|
---|
| 136 | el.elType = TextE;
|
---|
| 137 | while (c != '<' && !(compatMode && (c == PARA_TERM || c == REC_TERM))) {
|
---|
| 138 | el.text.push_back (c);
|
---|
| 139 | if (!buf.Next() || !buf.Peek (c)) break;
|
---|
| 140 | }
|
---|
| 141 |
|
---|
| 142 | return true;
|
---|
| 143 | }
|
---|
| 144 |
|
---|
| 145 | static void AddTextEl (TextElArray &doc, unsigned long &docLen,
|
---|
| 146 | const TextEl &el) {
|
---|
| 147 | doc.push_back (el);
|
---|
| 148 | docLen += el.text.size();
|
---|
| 149 |
|
---|
| 150 | // cout << "elType: " << el.elType << "\n";
|
---|
| 151 | // cout << "tagName: " << el.tagName << "\n";
|
---|
| 152 | // cout << "text: \"" << el.text << "\"\n\n";
|
---|
| 153 | }
|
---|
| 154 |
|
---|
| 155 |
|
---|
| 156 | bool ReadDoc (FileBuf &buf, const UCArray &docTag,
|
---|
| 157 | TextElArray &doc, unsigned long &docLen,
|
---|
| 158 | bool compatMode) {
|
---|
| 159 | bool compatInPara = false;
|
---|
| 160 | bool foundDocEl = false;
|
---|
| 161 | TextEl el;
|
---|
| 162 |
|
---|
| 163 | doc.erase (doc.begin(), doc.end());
|
---|
| 164 | docLen = 0;
|
---|
| 165 |
|
---|
| 166 | if (compatMode) {
|
---|
| 167 | // add <Document><Paragraph>
|
---|
| 168 | SetRecTag (el, OpenTagE);
|
---|
| 169 | AddTextEl (doc, docLen, el);
|
---|
| 170 | ToggleParaTag (el, compatInPara);
|
---|
| 171 | AddTextEl (doc, docLen, el);
|
---|
| 172 |
|
---|
| 173 | } else {
|
---|
| 174 | // look for an opening docTag
|
---|
| 175 | do {
|
---|
| 176 | if (!ReadTextEl (buf, el, compatMode, compatInPara)) return false;
|
---|
| 177 | } while (el.elType != OpenTagE || el.tagName != docTag);
|
---|
| 178 |
|
---|
| 179 | AddTextEl (doc, docLen, el);
|
---|
| 180 | }
|
---|
| 181 |
|
---|
| 182 |
|
---|
| 183 | // get all elements until the closing docTag
|
---|
| 184 | while (ReadTextEl (buf, el, compatMode, compatInPara)) {
|
---|
| 185 | foundDocEl = true;
|
---|
| 186 | AddTextEl (doc, docLen, el);
|
---|
| 187 | if (el.elType == CloseTagE && el.tagName == docTag)
|
---|
| 188 | return true;
|
---|
| 189 | }
|
---|
| 190 |
|
---|
| 191 |
|
---|
| 192 | if (compatMode) {
|
---|
| 193 | if (!foundDocEl) { // end of text
|
---|
| 194 | doc.erase (doc.begin(), doc.end());
|
---|
| 195 | return false;
|
---|
| 196 | }
|
---|
| 197 |
|
---|
| 198 | // if we got here then the eof was encountered before
|
---|
| 199 | // the closing document tag
|
---|
| 200 | if (compatInPara) {
|
---|
| 201 | // need to close this paragraph
|
---|
| 202 | ToggleParaTag (el, compatInPara);
|
---|
| 203 | AddTextEl (doc, docLen, el);
|
---|
| 204 | }
|
---|
| 205 | // close this document
|
---|
| 206 | SetRecTag (el, CloseTagE);
|
---|
| 207 | AddTextEl (doc, docLen, el);
|
---|
| 208 | }
|
---|
| 209 |
|
---|
| 210 | return true;
|
---|
| 211 | }
|
---|
| 212 |
|
---|
| 213 |
|
---|