| 2126 |
ghuddy |
1 |
using System;
|
|
|
2 |
using System.Collections;
|
|
|
3 |
using System.Text;
|
|
|
4 |
using Word;
|
| 2130 |
ghuddy |
5 |
using Microsoft.Office.Core;
|
| 2126 |
ghuddy |
6 |
|
| 2130 |
ghuddy |
7 |
|
| 2126 |
ghuddy |
8 |
// TODO
|
|
|
9 |
//
|
| 2130 |
ghuddy |
10 |
// 1) Possible let user and a tag simply with </>
|
| 2126 |
ghuddy |
11 |
|
|
|
12 |
|
|
|
13 |
|
|
|
14 |
namespace EA_DocGen
|
|
|
15 |
{
|
|
|
16 |
/// <summary>
|
|
|
17 |
/// Summary description for TextParser.
|
|
|
18 |
/// </summary>
|
| 2130 |
ghuddy |
19 |
public class TextParser
|
|
|
20 |
{
|
| 2126 |
ghuddy |
21 |
// An embedded formatting tag can be one of two types
|
|
|
22 |
// * An EA_DocGen special format
|
|
|
23 |
// * An MS-Word Style based format
|
|
|
24 |
private enum style_type_e
|
|
|
25 |
{
|
| 2130 |
ghuddy |
26 |
STYLE_EA_DOCGEN = 0, // style is an EA_DocGen specific style or pattern
|
|
|
27 |
STYLE_MS_WORD, // style is an MS-WORD style of character/font attribute
|
| 2126 |
ghuddy |
28 |
STYLE_UNKNOWN = -1
|
|
|
29 |
};
|
| 2130 |
ghuddy |
30 |
|
|
|
31 |
private enum style_handling_e
|
|
|
32 |
{
|
|
|
33 |
STYLE_POST_FORMAT, // text will require formatting AFTER the entire string has been serialised.
|
|
|
34 |
STYLE_USE_TAG_STYLE, // text will require the tag-implied style to be applied to it.
|
|
|
35 |
STYLE_USE_CALLER_STYLE // text will require the callers specified style to be applied to it.
|
|
|
36 |
};
|
|
|
37 |
|
|
|
38 |
// IDs for all of the styles we will process.
|
|
|
39 |
// Note that the arrangement of enum values in the following type, has been made to allow
|
|
|
40 |
// for some simple comparison tests during parsing, so do not re-arrange these enums without
|
|
|
41 |
// addressing the necessary code changes in the parsing function.
|
|
|
42 |
private enum style_id_e
|
|
|
43 |
{
|
|
|
44 |
// EA_DocGen styles (0..99)
|
|
|
45 |
STYLE_ID_TABLE = 0,
|
|
|
46 |
// add more EA_DocGen styles here when we need to
|
|
|
47 |
|
|
|
48 |
// MS-WORD styles (100..)
|
|
|
49 |
STYLE_ID_BODY1 = 100,
|
|
|
50 |
STYLE_ID_BOLD,
|
|
|
51 |
STYLE_ID_ITALIC,
|
|
|
52 |
STYLE_ID_UNDERLINE,
|
|
|
53 |
STYLE_ID_NORMAL,
|
|
|
54 |
STYLE_ID_WARNING,
|
|
|
55 |
STYLE_ID_NOTE,
|
|
|
56 |
STYLE_ID_CODETEXT,
|
|
|
57 |
// add more MS-WORD styles here when we need to
|
|
|
58 |
|
|
|
59 |
// ALL bullet/list styles must be >= 200 for easy detection during post-formatting
|
|
|
60 |
STYLE_ID_LIST_BULLET_0 = 200,
|
|
|
61 |
STYLE_ID_LIST_BULLET_1,
|
|
|
62 |
STYLE_ID_LIST_BULLET_2,
|
|
|
63 |
STYLE_ID_LIST_BULLET_3,
|
|
|
64 |
STYLE_ID_LIST_BULLET_4,
|
|
|
65 |
STYLE_ID_LIST_INDENT_0,
|
|
|
66 |
STYLE_ID_LIST_INDENT_1,
|
|
|
67 |
STYLE_ID_LIST_INDENT_2,
|
|
|
68 |
STYLE_ID_LIST_INDENT_3,
|
|
|
69 |
STYLE_ID_LIST_INDENT_4,
|
|
|
70 |
// Below here go lists that display numbering or lettering and so "restart numbering"
|
|
|
71 |
// attributes are important
|
|
|
72 |
STYLE_ID_LIST_NUMBER_0,
|
|
|
73 |
STYLE_ID_LIST_NUMBER_1,
|
|
|
74 |
STYLE_ID_LIST_NUMBER_2,
|
|
|
75 |
STYLE_ID_LIST_NUMBER_3,
|
|
|
76 |
STYLE_ID_LIST_NUMBER_4,
|
|
|
77 |
STYLE_ID_LIST_ALPHA_0,
|
|
|
78 |
STYLE_ID_LIST_ALPHA_1,
|
|
|
79 |
STYLE_ID_LIST_ALPHA_2,
|
|
|
80 |
STYLE_ID_LIST_ALPHA_3,
|
|
|
81 |
STYLE_ID_LIST_ALPHA_4
|
|
|
82 |
// do not put anything below here
|
|
|
83 |
};
|
|
|
84 |
|
| 2126 |
ghuddy |
85 |
// A token is a block of text associated with a tag (style) name and type
|
|
|
86 |
private struct token_type
|
|
|
87 |
{
|
| 2130 |
ghuddy |
88 |
public string txt; // This is the text content enclosed by the tag
|
|
|
89 |
public style_type_e styleType; // The type MS-WORD or EA_DocGen
|
|
|
90 |
public string styleName; // The name of the style
|
|
|
91 |
public style_id_e styleId; // The id of the style
|
|
|
92 |
public int level; // level (for list items only)
|
|
|
93 |
public style_handling_e styleHandling; // Handling attribute for the style
|
| 2126 |
ghuddy |
94 |
};
|
|
|
95 |
|
| 2130 |
ghuddy |
96 |
// Some formatting has to be applied after ALL the text of a description has been
|
| 2126 |
ghuddy |
97 |
// appended to the word document. For this formatting, we need to remember the word
|
| 2130 |
ghuddy |
98 |
// ranges that denote the text and the style name to apply. So we do it in a list of
|
|
|
99 |
// the following item.
|
| 2126 |
ghuddy |
100 |
private struct postFormat_type
|
|
|
101 |
{
|
|
|
102 |
public postFormat_type(Word.Range wr, token_type tk)
|
|
|
103 |
{
|
|
|
104 |
m_wr = wr;
|
|
|
105 |
m_tk = tk;
|
|
|
106 |
}
|
|
|
107 |
public Word.Range m_wr;
|
|
|
108 |
public token_type m_tk;
|
|
|
109 |
};
|
|
|
110 |
|
|
|
111 |
// Use a hash table for recording the allowed tags and their attributes, facilitating rapid
|
|
|
112 |
// lookup during parsing.
|
|
|
113 |
private static Hashtable styleDefs = null;
|
|
|
114 |
|
|
|
115 |
|
|
|
116 |
/// <summary>
|
|
|
117 |
/// Class initialisation function
|
|
|
118 |
/// </summary>
|
|
|
119 |
public static void initialise()
|
|
|
120 |
{
|
|
|
121 |
// initialsie the hash table
|
|
|
122 |
styleDefs = new Hashtable();
|
|
|
123 |
|
|
|
124 |
// This list dictates what tags a user can use in the notes text of an EA element. When adding new items
|
| 2130 |
ghuddy |
125 |
// to this list, we may have to update the parsing function obviously.
|
| 2126 |
ghuddy |
126 |
|
| 2130 |
ghuddy |
127 |
// The hash key is the short tag name that end-users will use in their descriptions. Tags found in
|
| 2126 |
ghuddy |
128 |
// user text is matched to these keys, and the style definition if found can then be used.
|
|
|
129 |
|
|
|
130 |
// EA_DocGen tags
|
| 2130 |
ghuddy |
131 |
styleDefs.Add( EA_Constants.EA_DocGenTable, formStyleDef( style_type_e.STYLE_EA_DOCGEN, style_id_e.STYLE_ID_TABLE, EA_Constants.EA_DocGenTable, 0, style_handling_e.STYLE_USE_TAG_STYLE ) );
|
| 2126 |
ghuddy |
132 |
|
|
|
133 |
// MS-Word formatting tags
|
| 2130 |
ghuddy |
134 |
styleDefs.Add( "b", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_BOLD, EA_Constants.styleName_Bold , 0, style_handling_e.STYLE_POST_FORMAT) );
|
|
|
135 |
styleDefs.Add( "i", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_ITALIC, EA_Constants.styleName_Italic , 0, style_handling_e.STYLE_POST_FORMAT) );
|
|
|
136 |
styleDefs.Add( "u", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_UNDERLINE, EA_Constants.styleName_Underline , 0, style_handling_e.STYLE_POST_FORMAT) );
|
| 2126 |
ghuddy |
137 |
|
| 2130 |
ghuddy |
138 |
styleDefs.Add( "lb0", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_BULLET_0, EA_Constants.styleName_ListBullet0,1, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
139 |
styleDefs.Add( "lb1", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_BULLET_1, EA_Constants.styleName_ListBullet1,2, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
140 |
styleDefs.Add( "lb2", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_BULLET_2, EA_Constants.styleName_ListBullet2,3, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
141 |
styleDefs.Add( "lb3", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_BULLET_3, EA_Constants.styleName_ListBullet3,4, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
142 |
styleDefs.Add( "lb4", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_BULLET_4, EA_Constants.styleName_ListBullet4,5, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
143 |
styleDefs.Add( "ln0", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_NUMBER_0, EA_Constants.styleName_ListNumber0,1, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
144 |
styleDefs.Add( "ln1", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_NUMBER_1, EA_Constants.styleName_ListNumber1,2, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
145 |
styleDefs.Add( "ln2", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_NUMBER_2, EA_Constants.styleName_ListNumber2,3, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
146 |
styleDefs.Add( "ln3", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_NUMBER_3, EA_Constants.styleName_ListNumber3,4, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
147 |
styleDefs.Add( "ln4", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_NUMBER_4, EA_Constants.styleName_ListNumber4,5, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
148 |
styleDefs.Add( "li0", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_INDENT_0, EA_Constants.styleName_ListIndent0,1, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
149 |
styleDefs.Add( "li1", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_INDENT_1, EA_Constants.styleName_ListIndent1,2, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
150 |
styleDefs.Add( "li2", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_INDENT_2, EA_Constants.styleName_ListIndent2,3, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
151 |
styleDefs.Add( "li3", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_INDENT_3, EA_Constants.styleName_ListIndent3,4, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
152 |
styleDefs.Add( "li4", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_INDENT_4, EA_Constants.styleName_ListIndent4,5, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
153 |
styleDefs.Add( "la0", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_ALPHA_0, EA_Constants.styleName_AlphaList0, 1, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
154 |
styleDefs.Add( "la1", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_ALPHA_1, EA_Constants.styleName_AlphaList1, 2, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
155 |
styleDefs.Add( "la2", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_ALPHA_2, EA_Constants.styleName_AlphaList2, 3, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
156 |
styleDefs.Add( "la3", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_ALPHA_3, EA_Constants.styleName_AlphaList3, 4, style_handling_e.STYLE_POST_FORMAT ) );
|
|
|
157 |
styleDefs.Add( "la4", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_LIST_ALPHA_4, EA_Constants.styleName_AlphaList4, 5, style_handling_e.STYLE_POST_FORMAT ) );
|
| 2126 |
ghuddy |
158 |
|
| 2130 |
ghuddy |
159 |
styleDefs.Add( "code", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_CODETEXT, EA_Constants.styleName_CodeText, 0, style_handling_e.STYLE_USE_TAG_STYLE ) );
|
|
|
160 |
styleDefs.Add( "normal", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_NORMAL, EA_Constants.styleName_Normal , 0, style_handling_e.STYLE_USE_TAG_STYLE ) );
|
|
|
161 |
styleDefs.Add( "note", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_NOTE, EA_Constants.styleName_Note , 0, style_handling_e.STYLE_USE_TAG_STYLE ) );
|
|
|
162 |
styleDefs.Add( "warn", formStyleDef( style_type_e.STYLE_MS_WORD, style_id_e.STYLE_ID_WARNING, EA_Constants.styleName_Warning , 0, style_handling_e.STYLE_USE_TAG_STYLE ) );
|
| 2126 |
ghuddy |
163 |
}
|
|
|
164 |
|
|
|
165 |
/// <summary>
|
|
|
166 |
/// Helper for initialise() function
|
|
|
167 |
/// </summary>
|
|
|
168 |
/// <param name="styleName"></param>
|
| 2130 |
ghuddy |
169 |
/// <param name="styleName"></param>
|
| 2126 |
ghuddy |
170 |
/// <returns></returns>
|
| 2130 |
ghuddy |
171 |
private static token_type formStyleDef(style_type_e styleType, style_id_e styleId, string styleName, int levelvar, style_handling_e styleHandling)
|
| 2126 |
ghuddy |
172 |
{
|
|
|
173 |
token_type tokenType = new token_type();
|
| 2130 |
ghuddy |
174 |
tokenType.styleType = styleType; // The kind of style (ms-word or ea_docgen)
|
|
|
175 |
tokenType.styleId = styleId; // The specific type of style
|
|
|
176 |
tokenType.styleName = styleName; // The name of the style
|
|
|
177 |
tokenType.txt = null; // This is the actual text to be formatted
|
|
|
178 |
tokenType.level = levelvar; // used to record level numbering - only really useful for bullet/list styles
|
|
|
179 |
tokenType.styleHandling = styleHandling;
|
| 2126 |
ghuddy |
180 |
return tokenType;
|
|
|
181 |
}
|
|
|
182 |
|
|
|
183 |
|
| 2130 |
ghuddy |
184 |
private static void initialiseToken(out token_type token)
|
|
|
185 |
{
|
|
|
186 |
token.styleName = EA_Constants.styleName_Body1;
|
|
|
187 |
token.styleId = style_id_e.STYLE_ID_BODY1;
|
|
|
188 |
token.styleType = style_type_e.STYLE_MS_WORD;
|
|
|
189 |
token.styleHandling = style_handling_e.STYLE_USE_CALLER_STYLE;
|
|
|
190 |
token.txt = null;
|
|
|
191 |
token.level = 0;
|
|
|
192 |
}
|
| 2126 |
ghuddy |
193 |
|
| 2134 |
ghuddy |
194 |
|
|
|
195 |
private static string convert_EA7_1_RTF_ListTag(string s, string open, string close, string replacement_open, string replacement_close)
|
|
|
196 |
{
|
|
|
197 |
int pos;
|
|
|
198 |
int pos_2;
|
|
|
199 |
|
|
|
200 |
pos = s.IndexOf(open,0);
|
|
|
201 |
if (pos >= 0)
|
|
|
202 |
{
|
|
|
203 |
while (pos >= 0)
|
|
|
204 |
{
|
|
|
205 |
pos_2 = s.IndexOf(close,pos);
|
|
|
206 |
|
|
|
207 |
string seg = s.Substring(pos+4, pos_2 - (pos+4));
|
|
|
208 |
seg = seg.Replace("\t<li>", replacement_open);
|
|
|
209 |
seg = seg.Replace("</li>", replacement_close);
|
|
|
210 |
|
|
|
211 |
s = s.Substring(0, pos) + seg + s.Substring(pos_2+5, s.Length - (pos_2+5));
|
|
|
212 |
|
|
|
213 |
pos = s.IndexOf(open,0);
|
|
|
214 |
}
|
|
|
215 |
}
|
|
|
216 |
return s;
|
|
|
217 |
}
|
|
|
218 |
|
|
|
219 |
|
|
|
220 |
|
|
|
221 |
|
| 2126 |
ghuddy |
222 |
/// <summary>
|
|
|
223 |
/// Parse the notes of an element and use the results to form document content.
|
|
|
224 |
/// </summary>
|
|
|
225 |
/// <param name="theElement"></param>
|
| 2130 |
ghuddy |
226 |
/// <param name="callerStyle"></param>
|
| 2126 |
ghuddy |
227 |
/// <returns></returns>
|
| 2130 |
ghuddy |
228 |
public static bool parse(string s, int id, string callerStyle, float indent_pts, bool continuation)
|
| 2126 |
ghuddy |
229 |
{
|
|
|
230 |
int pos;
|
|
|
231 |
int pos_LeftBracket = 0;
|
|
|
232 |
int pos_RightBracket = 0;
|
|
|
233 |
int pos_tagName;
|
|
|
234 |
int pos_ContentStart;
|
|
|
235 |
int pos_ContentEnd;
|
|
|
236 |
bool lookingForTagEnd;
|
|
|
237 |
bool foundError = false;
|
|
|
238 |
Word.Range wr_body;
|
|
|
239 |
|
| 2134 |
ghuddy |
240 |
// Convert EA7.1 embedded RTF controls into EA_DocGen controls where possible
|
|
|
241 |
s = convert_EA7_1_RTF_ListTag(s, "<ol>\r\n", "</ol>", "<ln0>", "</ln0>");
|
|
|
242 |
s = convert_EA7_1_RTF_ListTag(s, "<ul>\r\n", "</ul>", "<lb0>", "</lb0>");
|
| 2130 |
ghuddy |
243 |
|
| 2134 |
ghuddy |
244 |
// In EAv7.1, Sparx allows users to embed RTF into notes text. RTF uses <> to enclose controls just as EA_DocGen
|
|
|
245 |
// does for its own controls which this function decodes. However, EA_DocGen controls are entered as text by
|
|
|
246 |
// human users and so EAv7.1 converts < and > chars into a form such that they do not look like RTF tags.
|
|
|
247 |
// We have to convert EA's escaping mechanism back into ascii text chars otherwise the parser wont work.
|
|
|
248 |
// There may be issues to resolve with this parser if EAv7.1 users have used any of EA's RTF editing features
|
|
|
249 |
// because if they do, the < and > chars there will not be escaped and the parser will find them and try to
|
|
|
250 |
// decode them. Most will probably not be decodable.
|
|
|
251 |
s = s.Replace("<","<");
|
|
|
252 |
s = s.Replace(">",">");
|
|
|
253 |
|
| 2130 |
ghuddy |
254 |
// Begin to construct a range that will eventually encompass ALL of the text we will serialize during the execution
|
|
|
255 |
// of this function. This is needed later only if the caller has specified a requirement style that needs to have
|
|
|
256 |
// global strikethrough or italicising applied to (all of) the text
|
|
|
257 |
object startLocation;
|
|
|
258 |
object endLocation;
|
|
|
259 |
startLocation = createWordDoc.WordDocument.Content.End - 1;
|
|
|
260 |
|
|
|
261 |
// Requirement element text must be indented according to the level number of the requirement tag. Caller
|
|
|
262 |
// passes in absolute indentation value but for some items such as bullets/lists, we need a relative adjustment
|
|
|
263 |
// since bullet/list styles have their own indentation settings and we only want to offset them rather than overwrite
|
|
|
264 |
// them.
|
|
|
265 |
// Get relative (to 2.5cm) indentation user has commanded. The 2.5cm mark is the standard point where body 1 text
|
|
|
266 |
// begins. At 2.5cm, the pts value is 70.866. We only indent, never outdent so end stop at 0.
|
| 2126 |
ghuddy |
267 |
float relative_indent_adjustment = 0;
|
|
|
268 |
if (indent_pts > 0)
|
|
|
269 |
{
|
|
|
270 |
relative_indent_adjustment = indent_pts - (float)70.866;
|
|
|
271 |
if (relative_indent_adjustment < 0)
|
|
|
272 |
relative_indent_adjustment = 0;
|
|
|
273 |
}
|
|
|
274 |
|
| 2130 |
ghuddy |
275 |
// A working variable and a list for completed tokens
|
| 2126 |
ghuddy |
276 |
token_type token;
|
|
|
277 |
ArrayList tokens = new ArrayList();
|
|
|
278 |
|
|
|
279 |
// default starting token - may be updated later
|
| 2130 |
ghuddy |
280 |
initialiseToken(out token);
|
| 2126 |
ghuddy |
281 |
|
| 2130 |
ghuddy |
282 |
|
| 2126 |
ghuddy |
283 |
lookingForTagEnd = false;
|
|
|
284 |
pos_ContentStart = 0;
|
|
|
285 |
|
| 2130 |
ghuddy |
286 |
// PARSING LOOP -
|
|
|
287 |
// Break up the input string into tokens that identify what kind of action is to be performed
|
|
|
288 |
// with the token text. The default is as seen just above - apply Body1 MS-Word style. However,
|
|
|
289 |
// if user has used formatting tags, the token style name, ID, and type will be updated accordingly.
|
|
|
290 |
// This must be done iteratively until we have exhausted the input string.
|
|
|
291 |
|
| 2126 |
ghuddy |
292 |
// look for a tag
|
|
|
293 |
pos = s.IndexOf("<", 0);
|
|
|
294 |
while ((pos >= 0) && (pos < s.Length))
|
|
|
295 |
{
|
| 2128 |
ghuddy |
296 |
if (createWordDoc.abortCreationThread)
|
|
|
297 |
return false;
|
|
|
298 |
|
| 2126 |
ghuddy |
299 |
// record position of tag
|
|
|
300 |
pos_LeftBracket = pos;
|
|
|
301 |
|
|
|
302 |
// tag name begins at the next char
|
|
|
303 |
pos_tagName = pos_LeftBracket + 1;
|
|
|
304 |
|
|
|
305 |
// Check if this is a closing tag
|
|
|
306 |
bool isEnding = false;
|
|
|
307 |
if (pos < (s.Length-1))
|
|
|
308 |
{
|
|
|
309 |
if (s[pos+1] == '/')
|
|
|
310 |
{
|
|
|
311 |
// skip past the / char
|
|
|
312 |
isEnding = true;
|
|
|
313 |
pos_tagName++;
|
|
|
314 |
}
|
|
|
315 |
}
|
| 2130 |
ghuddy |
316 |
|
| 2126 |
ghuddy |
317 |
// We found a possible tag, now figure out if this is one of the tags we recognise
|
|
|
318 |
bool found = false;
|
|
|
319 |
|
|
|
320 |
// look for the closing bracket of the tag
|
|
|
321 |
pos = s.IndexOf(">", pos+1);
|
|
|
322 |
if (pos >= 0)
|
|
|
323 |
{
|
| 2130 |
ghuddy |
324 |
// use hash table to identify the tag
|
| 2126 |
ghuddy |
325 |
found = styleDefs.Contains(s.Substring(pos_tagName, pos - pos_tagName));
|
|
|
326 |
}
|
|
|
327 |
else
|
|
|
328 |
{
|
|
|
329 |
// Cannot find any '>' so we should just exit the loop
|
|
|
330 |
break;
|
|
|
331 |
}
|
|
|
332 |
|
|
|
333 |
// if the tag was recognised
|
|
|
334 |
if (found)
|
|
|
335 |
{
|
|
|
336 |
// record position of the closing bracket of the tag
|
|
|
337 |
pos_RightBracket = pos;
|
|
|
338 |
|
|
|
339 |
// if this is an end tag, ie. </tagname>
|
|
|
340 |
if (isEnding)
|
|
|
341 |
{
|
|
|
342 |
pos_ContentEnd = pos_LeftBracket - 1; // not sure if we really need to compute pos_ContentEnd
|
| 2130 |
ghuddy |
343 |
|
| 2126 |
ghuddy |
344 |
// check for out of sequence error
|
|
|
345 |
if (!lookingForTagEnd)
|
|
|
346 |
{
|
|
|
347 |
if (!foundError)
|
|
|
348 |
{
|
|
|
349 |
Main.WriteOutput( string.Format("ERROR, Found out of sequence style tag ({0}), generated document text may be formatted incorrectly.", s.Substring(pos_LeftBracket, pos_RightBracket - pos_LeftBracket + 1)), id);
|
|
|
350 |
foundError = true;
|
|
|
351 |
}
|
|
|
352 |
}
|
|
|
353 |
else
|
|
|
354 |
{
|
|
|
355 |
// Check that the end tag matches the start tag
|
| 2130 |
ghuddy |
356 |
// NOTE: If we were to allow lazy end-tagging (ie using '</>' only) we would have to do away with this
|
|
|
357 |
// check.
|
| 2126 |
ghuddy |
358 |
token_type tt = ((token_type)styleDefs[s.Substring(pos_tagName, pos_RightBracket-pos_tagName)]);
|
| 2130 |
ghuddy |
359 |
if (token.styleType == tt.styleType &&
|
|
|
360 |
token.styleId == tt.styleId &&
|
|
|
361 |
token.styleName == tt.styleName)
|
| 2126 |
ghuddy |
362 |
{
|
|
|
363 |
// Update the token's text field and add the now complete token to our list
|
|
|
364 |
// for processing a little later on.
|
|
|
365 |
token.txt = s.Substring(pos_ContentStart, pos_LeftBracket - pos_ContentStart);
|
|
|
366 |
tokens.Add(token);
|
|
|
367 |
|
|
|
368 |
// re-initialise token for next tag search
|
| 2130 |
ghuddy |
369 |
initialiseToken(out token);
|
| 2126 |
ghuddy |
370 |
|
|
|
371 |
lookingForTagEnd = false;
|
|
|
372 |
|
|
|
373 |
pos_ContentStart = pos_RightBracket + 1;
|
|
|
374 |
}
|
|
|
375 |
else
|
|
|
376 |
{
|
|
|
377 |
// end tag does not seem to be the same as the starting tag, so ignore it
|
|
|
378 |
if (!foundError)
|
|
|
379 |
{
|
|
|
380 |
Main.WriteOutput(string.Format("ERROR, Found unmatched style tag ({0}), generated document text may be formatted incorrectly.", s.Substring(pos_LeftBracket, pos_RightBracket - pos_LeftBracket + 1)), id);
|
|
|
381 |
foundError = true;
|
|
|
382 |
}
|
|
|
383 |
}
|
|
|
384 |
}
|
|
|
385 |
}
|
|
|
386 |
else
|
|
|
387 |
{
|
|
|
388 |
// If there is content prior to now that has not been consumed, tokenise it now
|
|
|
389 |
if ((pos_LeftBracket - pos_ContentStart) > 0)
|
|
|
390 |
{
|
|
|
391 |
token.txt = s.Substring(pos_ContentStart, pos_LeftBracket - pos_ContentStart);
|
|
|
392 |
tokens.Add(token);
|
|
|
393 |
}
|
| 2130 |
ghuddy |
394 |
|
| 2126 |
ghuddy |
395 |
if (lookingForTagEnd)
|
|
|
396 |
{
|
|
|
397 |
if (!foundError)
|
|
|
398 |
{
|
|
|
399 |
Main.WriteOutput(string.Format("ERROR, Found nested style tag ({0}), generated document text may be formatted incorrectly.", s.Substring(pos_LeftBracket, pos_RightBracket - pos_LeftBracket + 1)), id);
|
|
|
400 |
foundError = true;
|
|
|
401 |
}
|
|
|
402 |
}
|
|
|
403 |
else
|
|
|
404 |
{
|
| 2130 |
ghuddy |
405 |
// update the token variable with this tags atributes from the hash table lookup. This overwrites
|
|
|
406 |
// the default values assigned when parsing began, or when we resumed parsing after dealing with
|
|
|
407 |
// the previous token found.
|
|
|
408 |
token_type lookupToken = ((token_type)styleDefs[s.Substring(pos_tagName, pos_RightBracket-pos_tagName)]);
|
|
|
409 |
token.styleId = lookupToken.styleId;
|
|
|
410 |
token.styleType = lookupToken.styleType;
|
|
|
411 |
token.styleName = lookupToken.styleName;
|
|
|
412 |
token.level = lookupToken.level;
|
|
|
413 |
token.styleHandling = lookupToken.styleHandling;
|
| 2126 |
ghuddy |
414 |
token.txt = null; // we dont know what the text content will be yet. This is obtained when we encounter the end tag
|
|
|
415 |
|
|
|
416 |
pos_ContentStart = pos_RightBracket + 1;
|
|
|
417 |
|
|
|
418 |
lookingForTagEnd = true;
|
|
|
419 |
}
|
|
|
420 |
}
|
|
|
421 |
}
|
|
|
422 |
else
|
|
|
423 |
{
|
|
|
424 |
// the tag was not recognised so for now we just treat it as if it were plain text and continue
|
|
|
425 |
}
|
|
|
426 |
|
|
|
427 |
// look for next tag
|
|
|
428 |
pos = s.IndexOf("<", pos);
|
|
|
429 |
|
|
|
430 |
} // end of the loop
|
|
|
431 |
|
|
|
432 |
|
|
|
433 |
// take care of the last token, if there is one
|
|
|
434 |
if (pos_ContentStart < s.Length)
|
|
|
435 |
{
|
|
|
436 |
// Update the token's text field
|
|
|
437 |
token.txt = s.Substring(pos_ContentStart, s.Length - pos_ContentStart);
|
|
|
438 |
tokens.Add(token);
|
|
|
439 |
}
|
|
|
440 |
|
|
|
441 |
if (lookingForTagEnd)
|
|
|
442 |
{
|
|
|
443 |
if (!foundError)
|
|
|
444 |
{
|
|
|
445 |
Main.WriteOutput(string.Format("ERROR, Found incomplete style tag ({0}), generated document text may be formatted incorrectly.", s.Substring(pos_LeftBracket, pos_RightBracket - pos_LeftBracket + 1)), id);
|
|
|
446 |
foundError = true;
|
|
|
447 |
}
|
|
|
448 |
}
|
|
|
449 |
|
| 2130 |
ghuddy |
450 |
// The way MS-Word works makes it necessary to do some formatting after we have serialised all of the text.
|
|
|
451 |
// So, we need another list. This will contain elements that have the token AND the word range object that we
|
|
|
452 |
// obtain when we serialise the text.
|
| 2126 |
ghuddy |
453 |
ArrayList postFormats = new ArrayList();
|
|
|
454 |
|
| 2130 |
ghuddy |
455 |
// SERIALISATION LOOP - Now process all the tokens we have found
|
|
|
456 |
int tt_i = 0;
|
|
|
457 |
for (tt_i = 0; tt_i < tokens.Count; tt_i++)
|
| 2126 |
ghuddy |
458 |
{
|
| 2130 |
ghuddy |
459 |
token_type tt = (token_type)tokens[tt_i];
|
|
|
460 |
|
| 2128 |
ghuddy |
461 |
if (createWordDoc.abortCreationThread)
|
|
|
462 |
return false;
|
|
|
463 |
|
| 2126 |
ghuddy |
464 |
if (tt.txt != null && tt.txt.Length > 0)
|
|
|
465 |
{
|
|
|
466 |
|
|
|
467 |
switch (tt.styleType)
|
|
|
468 |
{
|
|
|
469 |
case style_type_e.STYLE_EA_DOCGEN:
|
| 2130 |
ghuddy |
470 |
switch (tt.styleId)
|
| 2126 |
ghuddy |
471 |
{
|
| 2130 |
ghuddy |
472 |
case style_id_e.STYLE_ID_TABLE:
|
| 2126 |
ghuddy |
473 |
TabularContent.processTableElement(tt.txt, 0, indent_pts);
|
|
|
474 |
continuation = false;
|
|
|
475 |
|
|
|
476 |
// flag list numbering restart
|
|
|
477 |
postFormats.Add( new postFormat_type(null, tt) );
|
|
|
478 |
break;
|
|
|
479 |
|
|
|
480 |
default:
|
|
|
481 |
break;
|
|
|
482 |
}
|
|
|
483 |
break;
|
|
|
484 |
|
|
|
485 |
case style_type_e.STYLE_MS_WORD:
|
| 2130 |
ghuddy |
486 |
switch (tt.styleHandling)
|
| 2126 |
ghuddy |
487 |
{
|
| 2130 |
ghuddy |
488 |
case style_handling_e.STYLE_POST_FORMAT:
|
|
|
489 |
// Replace <br> with actual required characters, and use the caller supplied style when serialising
|
|
|
490 |
// the text
|
|
|
491 |
tt.txt = tt.txt.Replace("<br>","\r\n");
|
|
|
492 |
|
|
|
493 |
// Serialise the text, initially applying callers style but since post-formatting will be
|
|
|
494 |
// done later, the texts appearance will change from what this initially applied style
|
|
|
495 |
// implies.
|
|
|
496 |
wr_body = TextualContent.appendAndSelectText( tt.txt, callerStyle, continuation );
|
| 2126 |
ghuddy |
497 |
continuation = true;
|
| 2130 |
ghuddy |
498 |
|
|
|
499 |
while (wr_body.Characters.Last.Text.Equals("\r") || wr_body.Characters.Last.Text.Equals("\n"))
|
|
|
500 |
wr_body.End = wr_body.End - 1; // don't format the /n or \r char at the end - doing so causes wierd ms-word exceptions later on
|
| 2126 |
ghuddy |
501 |
postFormats.Add( new postFormat_type(wr_body, tt) );
|
|
|
502 |
break;
|
|
|
503 |
|
| 2130 |
ghuddy |
504 |
case style_handling_e.STYLE_USE_TAG_STYLE:
|
|
|
505 |
// Replace <br> with actual required characters, and use the caller supplied style when serialising
|
|
|
506 |
// the text
|
|
|
507 |
tt.txt = tt.txt.Replace("<br>","\r\n");
|
|
|
508 |
|
|
|
509 |
// Serialise the text, applying the tag's style
|
| 2126 |
ghuddy |
510 |
wr_body = TextualContent.appendAndSelectText( tt.txt, tt.styleName, continuation );
|
|
|
511 |
continuation = true;
|
| 2130 |
ghuddy |
512 |
|
|
|
513 |
// Indent according to callers specified amount
|
| 2126 |
ghuddy |
514 |
if (indent_pts > 0)
|
|
|
515 |
wr_body.ParagraphFormat.LeftIndent = indent_pts;
|
|
|
516 |
|
| 2130 |
ghuddy |
517 |
// flag list numbering restart if this is printable text.
|
|
|
518 |
if (tt.txt.Trim().Length > 0)
|
|
|
519 |
postFormats.Add( new postFormat_type(null, tt) );
|
| 2126 |
ghuddy |
520 |
break;
|
|
|
521 |
|
|
|
522 |
|
| 2130 |
ghuddy |
523 |
case style_handling_e.STYLE_USE_CALLER_STYLE:
|
|
|
524 |
// Replace <br> with actual required characters, and use the caller supplied style when serialising
|
|
|
525 |
// the text
|
|
|
526 |
tt.txt = tt.txt.Replace("<br>","\r\n");
|
|
|
527 |
|
|
|
528 |
// Serialise the text, applying callers style
|
|
|
529 |
wr_body = TextualContent.appendAndSelectText( tt.txt, callerStyle, continuation );
|
| 2126 |
ghuddy |
530 |
continuation = true;
|
| 2130 |
ghuddy |
531 |
|
|
|
532 |
// Indent according to callers specified amount
|
| 2126 |
ghuddy |
533 |
if (indent_pts > 0)
|
|
|
534 |
wr_body.ParagraphFormat.LeftIndent = indent_pts;
|
|
|
535 |
|
|
|
536 |
// flag list numbering restart if this is printable text.
|
|
|
537 |
if (tt.txt.Trim().Length > 0)
|
|
|
538 |
postFormats.Add( new postFormat_type(null, tt) );
|
|
|
539 |
break;
|
|
|
540 |
|
|
|
541 |
default:
|
|
|
542 |
break;
|
|
|
543 |
}
|
|
|
544 |
break;
|
|
|
545 |
|
|
|
546 |
default:
|
|
|
547 |
break;
|
|
|
548 |
}
|
|
|
549 |
}
|
| 2130 |
ghuddy |
550 |
} // end of serialisation loop
|
| 2126 |
ghuddy |
551 |
|
| 2130 |
ghuddy |
552 |
|
|
|
553 |
// POST-FORMATTING LOOP - Now apply post formatting commands to text already serialised in previous loop
|
| 2126 |
ghuddy |
554 |
int last_list_level = 0;
|
|
|
555 |
foreach (postFormat_type pf in postFormats)
|
|
|
556 |
{
|
|
|
557 |
object style;
|
|
|
558 |
|
| 2128 |
ghuddy |
559 |
if (createWordDoc.abortCreationThread)
|
|
|
560 |
return false;
|
|
|
561 |
|
| 2126 |
ghuddy |
562 |
// a null word range implies we must restart numbering for any lists
|
|
|
563 |
if (pf.m_wr == null)
|
|
|
564 |
{
|
|
|
565 |
last_list_level = 0;
|
|
|
566 |
}
|
|
|
567 |
else
|
|
|
568 |
{
|
| 2130 |
ghuddy |
569 |
switch (pf.m_tk.styleId)
|
| 2126 |
ghuddy |
570 |
{
|
| 2130 |
ghuddy |
571 |
case style_id_e.STYLE_ID_BOLD:
|
| 2126 |
ghuddy |
572 |
pf.m_wr.Select();
|
|
|
573 |
createWordDoc.WordApp.Selection.Range.Bold = 1;
|
|
|
574 |
last_list_level = 0;
|
|
|
575 |
break;
|
|
|
576 |
|
| 2130 |
ghuddy |
577 |
case style_id_e.STYLE_ID_ITALIC:
|
| 2126 |
ghuddy |
578 |
pf.m_wr.Select();
|
|
|
579 |
createWordDoc.WordApp.Selection.Range.Italic = 1;
|
|
|
580 |
last_list_level = 0;
|
|
|
581 |
break;
|
|
|
582 |
|
| 2130 |
ghuddy |
583 |
case style_id_e.STYLE_ID_UNDERLINE:
|
| 2126 |
ghuddy |
584 |
pf.m_wr.Select();
|
|
|
585 |
createWordDoc.WordApp.Selection.Range.Underline = Word.WdUnderline.wdUnderlineSingle;
|
|
|
586 |
last_list_level = 0;
|
|
|
587 |
break;
|
|
|
588 |
|
| 2130 |
ghuddy |
589 |
default:
|
|
|
590 |
// Handle bullets/lists
|
|
|
591 |
if (pf.m_tk.styleId >= style_id_e.STYLE_ID_LIST_BULLET_0)
|
|
|
592 |
{
|
|
|
593 |
style = pf.m_tk.styleName;
|
|
|
594 |
pf.m_wr.Select();
|
|
|
595 |
createWordDoc.WordApp.Selection.Range.set_Style(ref style);
|
| 2126 |
ghuddy |
596 |
|
| 2130 |
ghuddy |
597 |
// Figure out if we have to restart numbering
|
|
|
598 |
if (last_list_level < pf.m_tk.level)
|
|
|
599 |
{
|
|
|
600 |
// only need to restart numbering if this list displays numbering - bullets and simple
|
|
|
601 |
// indents do not, but alpha and numeric lists do. The style_id_e type has been organised to
|
|
|
602 |
// make this a simple test.
|
|
|
603 |
if (pf.m_tk.styleId >= style_id_e.STYLE_ID_LIST_NUMBER_0)
|
|
|
604 |
{
|
|
|
605 |
// To restart numbering, the only way to do it is to (re-)apply the list template to the
|
|
|
606 |
// selection with a "continue previous list" setting of false, and an "apply to"
|
|
|
607 |
// setting of "whole list".
|
|
|
608 |
object continuePreviousList = false;
|
|
|
609 |
object applyTo = Word.WdListApplyTo.wdListApplyToWholeList;
|
|
|
610 |
object defListBehavour = Word.WdDefaultListBehavior.wdWord10ListBehavior;
|
|
|
611 |
|
|
|
612 |
Word.ListTemplate lt = createWordDoc.WordApp.Selection.Range.ListFormat.ListTemplate;
|
| 2126 |
ghuddy |
613 |
|
| 2130 |
ghuddy |
614 |
createWordDoc.WordApp.Selection.Range.ListFormat.ApplyListTemplate(
|
|
|
615 |
lt, ref continuePreviousList, ref applyTo, ref defListBehavour);
|
|
|
616 |
}
|
|
|
617 |
}
|
|
|
618 |
|
|
|
619 |
// shift content right by relative indent adjustment we calculated earlier
|
|
|
620 |
if (relative_indent_adjustment > 0)
|
| 2126 |
ghuddy |
621 |
{
|
| 2130 |
ghuddy |
622 |
createWordDoc.WordApp.Selection.Range.ParagraphFormat.LeftIndent += relative_indent_adjustment;
|
| 2126 |
ghuddy |
623 |
}
|
|
|
624 |
|
| 2130 |
ghuddy |
625 |
last_list_level = pf.m_tk.level;
|
| 2126 |
ghuddy |
626 |
}
|
|
|
627 |
break;
|
|
|
628 |
}
|
|
|
629 |
}
|
| 2130 |
ghuddy |
630 |
} // end of post-formatting loop
|
|
|
631 |
|
|
|
632 |
// Special handling for Proposed and Rejected requirement sections - here the text must be italicies or
|
|
|
633 |
// struck through (see definition of the styleName_ReqPropBody and styleName_ReqRejBody styles in StyleContent.cs).
|
|
|
634 |
if (callerStyle.Equals(EA_Constants.styleName_ReqPropBody))
|
|
|
635 |
{
|
|
|
636 |
// Complete construction of a range that will encompass ALL of the text we will serialize during the execution
|
|
|
637 |
// of this function
|
|
|
638 |
endLocation = createWordDoc.WordDocument.Content.End;
|
|
|
639 |
Word.Range wr_total = createWordDoc.WordDocument.Range(ref startLocation, ref endLocation);
|
|
|
640 |
while (wr_total.Characters.Last.Text.Equals("\r") || wr_total.Characters.Last.Text.Equals("\n"))
|
|
|
641 |
wr_total.End = wr_total.End - 1; // don't format the \r\n char at the end - doing so causes wierd ms-word exceptions later on
|
|
|
642 |
|
|
|
643 |
// italicise
|
|
|
644 |
wr_total.Font.Italic = (int)MsoTriState.msoTrue;
|
| 2126 |
ghuddy |
645 |
}
|
| 2130 |
ghuddy |
646 |
else if (callerStyle.Equals(EA_Constants.styleName_ReqRejBody))
|
|
|
647 |
{
|
|
|
648 |
// Complete construction of a range that will encompass ALL of the text we will serialize during the execution
|
|
|
649 |
// of this function
|
|
|
650 |
endLocation = createWordDoc.WordDocument.Content.End;
|
|
|
651 |
Word.Range wr_total = createWordDoc.WordDocument.Range(ref startLocation, ref endLocation);
|
|
|
652 |
while (wr_total.Characters.Last.Text.Equals("\r") || wr_total.Characters.Last.Text.Equals("\n"))
|
|
|
653 |
wr_total.End = wr_total.End - 1; // don't format the \r\n char at the end - doing so causes wierd ms-word exceptions later on
|
|
|
654 |
|
|
|
655 |
// strikethrough - use msoCTrue since msoTrue simply toggles strikethough attribute, or so it seems
|
|
|
656 |
wr_total.Font.StrikeThrough = (int)MsoTriState.msoCTrue;
|
|
|
657 |
}
|
|
|
658 |
|
| 2126 |
ghuddy |
659 |
return true;
|
|
|
660 |
}
|
|
|
661 |
|
|
|
662 |
|
|
|
663 |
|
|
|
664 |
|
| 2130 |
ghuddy |
665 |
|
|
|
666 |
}
|
| 2126 |
ghuddy |
667 |
}
|