1 # Writer Application Code
3 Exact history was lost before Sept. 18th, 2000, but old source code
4 comments show that Writer core dates back until at least November
8 * `inc`: headers available to all source files inside the module
9 * `qa`: unit, slow and subsequent tests
12 * `uiconfig`: user interface configuration
13 * `util`: UNO passive registration config
16 * `core`: Writer core (document model, layout, UNO API implementation)
17 * `filter`: Writer internal filters
18 * `ascii`: plain text filter
20 * `docx`: wrapper for the UNO DOCX import filter (in writerfilter) for autotext purposes
22 * `inc`: include files for filters
23 * `rtf`: thin copy&paste helper around the UNO RTF import filter (in writerfilter)
25 * `ww8`: DOC import, DOC/DOCX/RTF export
26 * `xml`: ODF import/export, subclassed from xmloff (where most of the work is done)
27 * `uibase`: user interface (those parts that are linked into `sw` & always loaded)
28 * `ui`: user interface (optional parts that are loaded on demand (`swui`))
32 There is a good overview documentation of basic architecture of Writer core
35 - <https://wiki.openoffice.org/wiki/Writer/Core_And_Layout>
36 - <https://wiki.openoffice.org/wiki/Writer/Text_Formatting>
38 Writer specific WhichIds are defined in `sw/inc/hintids.hxx`.
40 The details below are mainly about details missing from the wiki pages.
44 The central class for a document is `SwDoc`, which represents a document.
46 A lot of the functionality is split out into separate Manager classes,
47 each of which implements some `IDocument*` interface; there are
48 `SwDoc::getIDocument*()` methods to retrieve the managers.
50 However there are still too many members and methods in this class,
51 many of which could be moved to some Manager or other...
55 Basically a (fancy) array of `SwNode` pointers. There are special subclasses of
56 `SwNode` (`SwStartNode` and `SwEndNode`) which are used to encode a nested tree
57 structure into the flat array; the range of nodes from `SwStartNode` to its
58 corresponding `SwEndNode` is sometimes called a "section" (but is not necessarily
59 what the high-level document model calls a "Section"; that is just one of the
62 The `SwNodes` contains the following top-level sections:
66 3. Frame / Header / Footer content
67 4. Deleted Change Tracking content
72 The Undo/Redo information is stored in a `sw::UndoManager` member of `SwDoc`,
73 which implements the `IDocumentUndoRedo` interface.
74 Its members include a `SwNodes` array containing the document content that
75 is currently not in the actual document but required for Undo/Redo, and
76 a stack of `SwUndo` actions, each of which represents one user-visible
79 There are also `ListActions` which internally contain several individual `SwUndo`
80 actions; these are created by the StartUndo/EndUndo wrapper methods.
84 The sub-structure of paragraphs is stored in the `SwpHintsArray` member
85 `SwTextNode::m_pSwpHints`. There is a base class `SwTextAttr` with numerous
86 subclasses; the `SwTextAttr` has a start and end index and a `SfxPoolItem`
87 to store the actual formatting attribute.
89 There are several sub-categories of `SwTextAttr`:
91 - formatting attributes: Character Styles (`SwTextCharFormat`, `RES_TXTATR_CHARFMT`)
92 and Automatic Styles (no special class, `RES_TXTATR_AUTOFMT`):
93 these are handled by `SwpHintsArray::BuildPortions` and MergePortions,
94 which create non-overlapping portions of formatting attributes.
96 - nesting attributes: Hyperlinks (`SwTextINetFormat`, `RES_TXTATR_INETFMT`),
97 Ruby (`SwTextRuby`, `RES_TXTATR_CJK_RUBY`) and Meta/MetaField (`SwTextMeta`,
98 `RES_TXTATR_META/RES_TXTATR_METAFIELD`):
99 these maintain a properly nested tree structure.
100 The Meta/Metafield are "special" because they have both start/end
101 and a dummy character at the start.
103 - misc. attributes: Reference Marks, ToX Marks
105 - attributes without end: Fields, Footnotes, Flys (`AS_CHAR`)
106 These all have a corresponding dummy character in the paragraph text, which
107 is a placeholder for the "expansion" of the attribute, e.g. field content.
111 There are multiple model classes involved for fields:
113 - `enum SwFieldIds` enumerates the different types of fields.
114 - `SwFieldType` contains some shared stuff for all fields of a type.
115 There are many subclasses of `SwFieldType`, one for each different type
117 For most types of fields there is one shared instance of this per type,
118 which is created in `DocumentFieldsManager::InitFieldTypes()`
119 but for some there are more than one, and they are dynamically created, see
120 `DocumentFieldsManager::InsertFieldType()`. An example for the latter are
121 variable fields (`SwFieldIds::GetExp/SwFieldIds::SetExp`), with one `SwFieldType` per
123 - `SwXFieldMaster` is the UNO wrapper of a field type.
124 It is a `SwClient` registered at the `SwFieldType`.
125 Its life-cycle is determined by UNO clients outside of `sw`; it will get
126 disposed when the `SwFieldType` dies.
127 - `SwFormatField` is the `SfxPoolItem` of a field.
128 The `SwFormatField` is a `SwClient` registered at its `SwFieldType`.
129 The `SwFormatField` owns the `SwField` of the field.
130 - `SwField` contains the core logic of a field.
131 The `SwField` is owned by the `SwFormatField` of the field.
132 There are many subclasses of `SwField`, one for each different type of field.
133 Note that there are not many places that can Expand the field to its
134 correct value, since for example page number fields require a View
135 with an up to date layout; therefore the correct expansion is cached.
136 - `SwTextField` is the text attribute of a field.
137 It owns the `SwFormatField` of the field (like all text attributes).
138 - `SwXTextField` is the UNO wrapper object of a field.
139 It is a `SwClient` registered at the `SwFormatField`.
140 Its life-cycle is determined by UNO clients outside of `sw`; it will get
141 disposed when the `SwFormatField` dies.
145 - `SwNumFormat` (subclass of `SvxNumFormat`) determines the formatting of a single
148 - `SwNumRule` (NOT a subclass of `SvxNumRule`) is a *list style*, containing one
149 `SwNumFormat` per list level.
150 `SwNumRule::maTextNodeList` is the list of `SwTextNode` that have this list style
153 - `SwNumberTreeNode` is a base class that represents an abstract node in a
154 hierarchical tree of numbered nodes.
156 - `SwNodeNum` is the subclass of `SwNumberTreeNode` that connects it with an
157 actual `SwTextNode` and also with a `SwNumRule`;
158 `SwTextNode::mpNodeNum` points back in the other direction
160 - `SwList` represents a list, which is (mostly) a vector of `SwNodeNum` trees,
161 one per `SwNodes` top-level section (why that?).
163 - `IDocumentListsAccess`, `sw::DocumentListsManager` owns all `SwList` instances,
164 and maintains mappings:
165 + from list-id to `SwList`
166 + from list style name to `SwList` (the "default" `SwList` for that list style)
168 - `IDocumentListItems`, `sw::DocumentListItemsManager` contains a set of all
169 `SwNodeNum` instances, ordered by `SwNode` index
171 - the special Outline numbering rule: `SwDoc::mpOutlineRule`
173 - `IDocumentOutlineNodes`, `sw::DocumentOutlineNodesManager` maintain
174 a list (which is actually stored in `SwNodes::m_pOutlineNodes`) of `SwTextNodes`
175 that either have the Outline numrule applied,
176 or have the `RES_PARATR_OUTLINELEVEL` item set (note that in the latter case,
177 the `SwTextNode` does not have a `SwNodeNum` and is not associated with the
178 `SwDoc::mpOutlineRule`).
180 - `SwTextNodes` and paragraph styles have items/properties:
181 + `RES_PARATR_OUTLINELEVEL/"OutlineLevel"` to specify an outline level without
182 necessarily having the outline `SwNumRule` assigned
183 + `RES_PARATR_NUMRULE/"NumberingStyleName"` the list style to apply; may be
184 empty `""` which means no list style (to override inherited value)
185 Only `SwTextNode` has these items:
186 + `RES_PARATR_LIST_ID/"ListId"`
187 determines the `SwList` to which the node is added
188 + `RES_PARATR_LIST_LEVEL/"NumberingLevel"`
189 the level at which the `SwTextNode` will appear in the list
190 + `RES_PARATR_LIST_ISRESTART/"ParaIsNumberingRestart"`
191 restart numbering sequence at this `SwTextNode`
192 + `RES_PARATR_LIST_RESTARTVALUE/"NumberingStartValue"`
193 restart numbering sequence at this `SwTextNode` with this value
194 + `RES_PARATR_LIST_ISCOUNTED/"NumberingIsNumber"`
195 determines if the node is actually counted in the numbering sequence;
196 these are different from `"phantoms"` because there's still a `SwTextNode`.
198 Note that there is no UNO service to represent a list.
202 The layout is a tree of `SwFrame` subclasses, the following relationships are
203 possible between frames:
205 - You can visit the tree by following the upper, lower, next and previous pointers.
206 - The functionality of flowing of a frame across multiple parents (e.g. pages)
207 is implemented in `SwFlowFrame`, which is not an `SwFrame` subclass. The logical
208 chain of such frames can be visited using the follow and precede pointers.
209 ("Leaf" is a term that refers to such a relationship.)
210 - In case a frame is split into multiple parts, then the first one is called
211 master, while the others are called follows.