1 Introduction
2 ------------
3
4 XSLTools is a collection of modules and packages facilitating the development
5 of applications based on XML, XSL stylesheets and transformations, notably Web
6 applications involving complicated Web forms potentially consisting of
7 editable hierarchical structures and potentially involving "live" or "in-page"
8 dynamic updates to portions of those Web forms.
9
10 Quick Start
11 -----------
12
13 Try running the demo:
14
15 python tools/demo.py
16
17 An introductory guide to creating applications can be found in the docs
18 directory - see docs/index.html for the start page.
19
20 Contact, Copyright and Licence Information
21 ------------------------------------------
22
23 The current Web page for XSLTools at the time of release is:
24
25 http://www.boddie.org.uk/python/XSLTools.html
26
27 Copyright and licence information can be found in the docs directory - see
28 docs/COPYING.txt, docs/lgpl-3.0.txt, docs/gpl-3.0.txt,
29 docs/LICENCE-Sarissa-lgpl.txt and docs/LICENCE-Sarissa-gpl.txt for more
30 information.
31
32 Additional Software
33 -------------------
34
35 The XSLTools examples make use of sarissa.js from the Sarissa distribution,
36 release 0.9.8.1. Copies of this file are found in the Resources/scripts
37 directory within each example's package.
38
39 Dependencies
40 ------------
41
42 XSLTools has the following basic dependencies:
43
44 Packages Release Information
45 -------- -------------------
46
47 libxml2dom 0.4.3
48 libxml2 and libxslt Some combinations may not be reliable!
49 Tested with libxml2 2.6.17 and libxslt 1.1.12
50 Tested with libxml2 2.6.27 and libxslt 1.1.20
51 libxslt 1.1.17 should be avoided
52
53 The example Web applications require WebStack (release 1.2.3 or later).
54 The example PyQt applications have been tested with PyQt 3.15.
55
56 New in XSLTools 0.5 (Changes since XSLTools 0.4.6)
57 --------------------------------------------------
58
59 * Renamed template:id to template:section in order to get around problems
60 and potential future problems with browsers (such as Konqueror 3.5.6)
61 which get confused loading documents with multiple id attributes, even if
62 only one is unprefixed. WARNING! This change is not backwards compatible.
63 * Added Ubuntu Feisty (7.04) package support.
64 * Tidied the documentation HTML.
65
66 New in XSLTools 0.4.6 (Changes since XSLTools 0.4.5)
67 ----------------------------------------------------
68
69 * Improved the Login module, enabling the VerySimpleWithLogin example for
70 various frameworks.
71 * Relicensed under the LGPL version 3 or later.
72 * Upgraded to Sarissa 0.9.7.8 (compatible with LGPL/GPL version 3), removing
73 a test around DOMParser so that Konqueror 3.4.0 is still supported.
74
75 New in XSLTools 0.4.5 (Changes since XSLTools 0.4.4)
76 ----------------------------------------------------
77
78 * Fixed the result of transformations in XSLOutput: proper document nodes
79 are now produced.
80 * Added an XSLForms.Resources.Login module which provides resources to
81 support login screens and redirects.
82 * Fixed newlines in the attributes created from fields in XSLForms: CR
83 characters are no longer included since this caused the doubling up of
84 newlines in Firefox.
85
86 New in XSLTools 0.4.4 (Changes since XSLTools 0.4.3)
87 ----------------------------------------------------
88
89 * Fixed translation selection for the template:i18n annotation attribute,
90 not just for the template:i18n extension function - more apologies for
91 resulting output changes!
92 * Improved the template fixing stylesheet and added some documentation for
93 the script and the related expr-prefix attribute.
94 * Introduced WebStack 1.2.2 EncodingSelector and encoding changes.
95 * Added docstring and return value for the write_month_to_document function
96 in XSLTools.XMLCalendar.
97
98 New in XSLTools 0.4.3 (Changes since XSLTools 0.4.2)
99 ----------------------------------------------------
100
101 * Fixed translation selection when an unsupported locale is specified,
102 choosing the first locale as the default (rather than exposing the values
103 themselves as translations). Note that this is an unfortunate and subtle
104 change which may affect application output - apologies are hereby offered!
105
106 New in XSLTools 0.4.2 (Changes since XSLTools 0.4.1)
107 ----------------------------------------------------
108
109 * Added a content type check in the XSLFormsResource class, permitting
110 non-form-based resources to access the raw request data, rather than have
111 the data processed unsuccessfully and consequently discarded.
112 * Added a script and a function to fix template namespaces after editing in
113 a careless editor.
114 * Changed libxml2mod and libxsltmod import details to try libxmlmods -
115 suggested by Lucian Wischik for libxml2dom.
116
117 New in XSLTools 0.4.1 (Changes since XSLTools 0.4)
118 --------------------------------------------------
119
120 * Made translations specified using the template:i18n annotation take
121 priority over template:value annotations.
122 * Added expression-based template:i18n annotations, and provided fallback
123 output for such translations based on the value of the evaluated
124 expression.
125
126 New in XSLTools 0.4 (Changes since XSLTools 0.3.1)
127 --------------------------------------------------
128
129 * Changed the preparation of templates to produce rule-based output
130 stylesheets, thus permitting recursive templates. This requires an extra
131 expr-prefix annotation to be used in certain kinds of templates.
132 * Added a recursive template example application.
133 * Changed fragment production to use original template documents instead of
134 output stylesheets.
135 * Changed the in_page_resources attribute to provide the output identifier,
136 thus changing the prepare_fragment method in Web resources so that only
137 the fragment identifier needs to be supplied.
138 * Added the XSLForms.Resources.WebResources.prepare_resources method for the
139 preparation of initialiser and output stylesheets before an application is
140 run.
141 * Changed selectors to not automatically create elements in the form data
142 document unless requested to do so. Introduced a Form.get_selector
143 method in XSLForms.Fields.
144 * Permitted the creation of hierarchies of elements in
145 XSLForms.Utils.add_elements.
146 * Introduced dynamic parameter evaluation for multiple-choice fields in
147 order to support sources of multiple-choice values which reside in the
148 form data document itself.
149 * Added the FixNamespace.xsl stylesheet to correct documents saved by HTML
150 editors which strip namespace prefixes.
151 * Fixed filesystem encoding issues in the Candidate example; fixed language
152 preference access in the Configurator and VerySimple examples.
153 * Changed the BaseHTTPRequestHandler version of the Candidate example to
154 store data in a subdirectory of the current working directory, thus
155 allowing the demonstration application to work after package installation.
156
157 New in XSLTools 0.3.1 (Changes since XSLTools 0.3)
158 --------------------------------------------------
159
160 * Fixed copyright and licensing information.
161
162 New in XSLTools 0.3 (Changes since XSLTools 0.2)
163 ------------------------------------------------
164
165 * Introduced copying of multiple-choice value element contents so that
166 option element labels can differ from the underlying values.
167 * Added internationalisation support, providing the template:i18n annotation
168 and the template:i18n extension function.
169 * Updated the documentation to cover the above new features.
170 * Fixed non-GET/POST request method handling in WebResources.
171 * Added the xslform_preparemacro.py script.
172 * Added an experimental template:range extension function.
173
174 New in XSLTools 0.2 (Changes since XSLTools 0.1)
175 ------------------------------------------------
176
177 * Made a new XSLTools package and moved XSLOutput into it.
178 * Improved serialisation of transformation results so that output options
179 are observed (in some cases, at least).
180 * Fixed stylesheet and reference document paths so that libxslt should not
181 now become confused by ambiguous relative paths.
182 * Added expression parameters to XSLOutput.Processor so that in-document
183 data can be used to, for example, initialise multiple-choice field values.
184 * Added input/initialiser support so that input documents can be tidied or
185 initialised using information from the template.
186 * Added template:init for use with template:element in XSLForms to control
187 element initialisation where necessary.
188 * Added special high-level "macro" attributes (eg. template:attribute-field)
189 which should make templates easier to write and maintain.
190 * Added template:if to XSLForms, providing conditional output of annotated
191 elements.
192 * Added set_document to XSLForms.Fields.Form.
193 * Added prepare_parameters to the XSLFormsResource class in the
194 XSLForms.Resources.WebResources module.
195 * Added element-path, url-encode and choice XSLForms extension functions.
196 * Improved Unicode support in the XSLForms extension functions.
197 * Changed in-page requests to contain proper POST data.
198 * Fixed checkbox and radiobutton value detection in XSLForms.js.
199 * Updated the code to work with WebStack 1.0 changes and adopted the
200 new-style WebStack demonstration mechanism.
201 * Added XMLCalendar and XMLTable (to the XSLTools package).
202 * Added a dictionary (or word lookup) example application.
203 * Added a job candidate profile (or CV editor) example application.
204 * Added a template attribute reference and an XSLFormsResource guide to the
205 documentation.
206 * Added Debian package support (specifically Ubuntu package support).
207 * Added missing COPYING.txt file.
208 * Renamed the scripts to avoid naming issues in system-wide installations.
209 * Added a PyQt example based on the system configurator example, with the
210 form prepared in Qt Designer. This example runs in PyQt and in a Web
211 environment without any changes to the application code. In-page updates
212 are currently not implemented in the Web version, however.
213
214 Notes on In-Page Update Functionality
215 -------------------------------------
216
217 Special note #1: Konqueror seems in certain cases to remember replaced form
218 content (when replaceChild is used to replace regions of the page which
219 include form elements). This causes the browser to believe that more form
220 fields exist on the page than actually do so, and subsequent form submissions
221 thus include the values of such removed fields. A special hack is in place to
222 disable form fields by changing their names, thus causing Konqueror to not
223 associate such fields with the real, active fields; this hack does not seem to
224 cause problems for Mozilla. This needs some investigation to determine in
225 exactly which circumstances the problem arises.
226
227 Special note #2: Konqueror also seems to crash if asked to find elements using
228 an empty 'id' attribute string. This needs some investigation to see if it
229 really is the getElementById call that causes the crash.
230
231 Special note #3: Konqueror's XMLHttpRequest seems to append null characters to
232 the end of field values. Attempting to prune them before the request is sent
233 fails with a function like the following:
234
235 function fixValue(fieldValue) {
236 if (fieldValue.length == 0) {
237 return fieldValue;
238 } else if (fieldValue[fieldValue.length - 1] == '\0') {
239 return fieldValue.substr(0, fieldValue.length - 1);
240 } else {
241 return fieldValue;
242 }
243 }
244
245 This may be because it is the entire message that is terminated with the null
246 character, and that this happens only upon sending the message. Consequently,
247 some frameworks (notably mod_python) do not support in-page functionality when
248 used from Konqueror.
249
250 Various browsers (eg. Mozilla/Firefox, Konqueror) will not allow the
251 XMLHttpRequest in-page updates to function unless the URL used in the
252 requestUpdate JavaScript function is compatible with the URL at which the
253 browser finds the application. Currently, relative URLs are in use to avoid
254 this issue of compatibility, but should an absolute URL be deduced using the
255 WebStack API and then used, it may be possible that the values returned by
256 that API do not match the actual addresses entered into the address bar of the
257 browser.
258
259 To check the behaviour of the applications, it is possible to view the
260 document source of the pages served by applications and to verify that the
261 URLs mentioned in the JavaScript function calls (to 'requestUpdate') either be
262 a relative link or involve a URL similar to that which appears in the
263 browser's address bar. In some environments, the use of 'localhost' addresses
264 often confuses the browser and server; one workaround is to use real host
265 names or addresses instead of 'localhost'.
266
267 Choosing an element-path:
268
269 When specifying the "context" of the in-page update, one must imagine which
270 element the template fragment should operate within. If the template:id
271 attribute marks a particular section, then the element-path should be a path
272 to the applicable context element for that section in the complete template
273 document. Note that if a template:element attribute appears on the same
274 element as the template:id attribute then the element-path should refer to the
275 element specified in the template:element attribute.
276
277 Choosing where to put template:attribute, template:id and id:
278
279 When specifying the extent of a template fragment, one must be sure not to put
280 the template:id attribute on the same element as a template:attribute
281 annotation; otherwise, the generated code will be improperly extracted as a
282 fragment producing two versions of the element - one for when the specified
283 attribute is present, and one for when it is not present. Generally,
284 template:id and id can be placed on the same node, however.
285
286 Stable element ordering and element-path:
287
288 Within the element-path, the numbering of the elements will start at 1.
289 Therefore it is vital to choose a region of the form data structure with the
290 element-path which is isolated from surrounding elements whose positions would
291 otherwise be dependent on a stable ordering of elements, and whose processing
292 would be disrupted if some new elements suddenly appeared claiming the same
293 positions in the document. For example:
294
295 <item value=""> .../item$1/value
296 <type value=""/> .../item$1/type$1/value
297 <comment value=""/> .../item$1/comment$2/value
298 </item>
299
300 In-page update...
301
302 <comment value=""/> .../item$1/comment$1/value
303
304 Notes on XSL
305 ------------
306
307 libxslt seems to be quite liberal on the definition of runtime parameters, in
308 that there is no apparent need to explicitly declare the corresponding global
309 variables in stylesheets. Whilst this is nice, we may eventually need to
310 detect such variables and add them in the preparation process.
311
312 Release Procedures
313 ------------------
314
315 Update the XSLTools/__init__.py and XSLForms/__init__.py __version__
316 attributes.
317 Change the version number and package filename/directory in the documentation.
318 Change code examples in the documentation if appropriate.
319 Update the release notes (see above).
320 Check the setup.py file and ensure that all package directories are mentioned.
321 Check the release information in the PKG-INFO file and in the package
322 changelog (and other files).
323 Tag, export.
324 Generate the example resources.
325 Generate the API documentation.
326 Remove generated .pyc files: rm `find . -name "*.pyc"`
327 Archive, upload.
328 Upload the introductory documentation.
329 Update PyPI, PythonInfo Wiki, Vaults of Parnassus entries.
330
331 Generating the Example Resources
332 --------------------------------
333
334 In order to prepare the example resources, the prepare_demo.py script must be
335 run as follows:
336
337 python tools/prepare_demo.py
338
339 This will ensure that all initialiser and output stylesheets are created and
340 are thus installed by packages.
341
342 Generating the API Documentation
343 --------------------------------
344
345 In order to prepare the API documentation, it is necessary to generate some
346 Web pages from the Python source code. For this, the epydoc application must
347 be available on your system. Then, inside the distribution directory, run the
348 apidocs.sh tool script as follows:
349
350 ./tools/apidocs.sh
351
352 Some warnings may be generated by the script, but the result should be a new
353 apidocs directory within the distribution directory.
354
355 Making Packages
356 ---------------
357
358 To make Debian-based packages:
359
360 1. Create new package directories under packages if necessary.
361 2. Make a symbolic link in the distribution's root directory to keep the
362 Debian tools happy:
363
364 ln -s packages/ubuntu-hoary/python2.4-xsltools/debian/
365
366 3. Run the package builder:
367
368 dpkg-buildpackage -rfakeroot
369
370 4. Locate and tidy up the packages in the parent directory of the
371 distribution's root directory.