0024526: Guide on Automatic Test System is obsolete
[occt.git] / dox / dev_guides / tests / tests.md
72b7576f 1 Automated Testing System {#dev_guides__tests}
e5bd0d98 4@tableofcontents
72b7576f 6@section testmanual_1 Introduction
504a8968 8This document provides OCCT developers and contributors with an overview and practical guidelines for work with OCCT automatic testing system.
10Reading the Introduction is sufficient for OCCT developers to use the test system to control non-regression of the modifications they implement in OCCT. Other sections provide a more in-depth description of the test system, required for modifying the tests and adding new test cases.
72b7576f 11
12@subsection testmanual_1_1 Basic Information
504a8968 14OCCT automatic testing system is organized around DRAW Test Harness @ref user_guides__test_harness "DRAW Test Harness", a console application based on Tcl (a scripting language) interpreter extended by OCCT-related commands.
e5bd0d98 15
504a8968 16Standard OCCT tests are included with OCCT sources and are located in subdirectory *tests* of the OCCT root folder. Other test folders can be included in the test system, e.g. for testing applications based on OCCT.
e5bd0d98 17
504a8968 18The tests are organized in three levels:
72b7576f 19
504a8968 20 * Group: a group of related test grids, usually testing a particular OCCT functionality (e.g. blend);
21 * Grid: a set of test cases within a group, usually aimed at testing some particular aspect or mode of execution of the relevant functionality (e.g. buildevol);
22 * Test case: a script implementing an individual test (e.g. K4).
24See <a href="#testmanual_5_1">Test Groups</a> for the current list of available test groups and grids.
72b7576f 25
504a8968 26Some tests involve data files (typically CAD models) which are located separately and are not included with OCCT code. The archive with publicly available test data files should be downloaded and installed independently on OCCT sources (from http://dev.opencascade.org).
72b7576f 27
28@subsection testmanual_1_2 Intended Use of Automatic Tests
30Each modification made in OCCT code must be checked for non-regression
31by running the whole set of tests. The developer who does the modification
504a8968 32is responsible for running and ensuring non-regression for the tests available to him.
34Note that many tests are based on data files that are confidential and thus available only at OPEN CASCADE. Thus official certification testing of the changes before integration to master branch of official OCCT Git repository (and finally to the official release) is performed by OPEN CASCADE in any case.
36Each new non-trivial modification (improvement, bug fix, new feature) in OCCT should be accompanied by a relevant test case suitable for verifying that modification. This test case is to be added by the developer who provides the modification.
72b7576f 37
504a8968 38If a modification affects the result of an existing test case, either the modification should be corrected (if it causes regression) or the affected test cases should be updated to account for the modification.
72b7576f 39
504a8968 40The modifications made in the OCCT code and related test scripts should be included in the same integration to the master branch.
72b7576f 41
42@subsection testmanual_1_3 Quick Start
44@subsubsection testmanual_1_3_1 Setup
504a8968 46Before running tests, make sure to define environment variable *CSF_TestDataPath* pointing to the directory containing test data files.
47(Publicly available data files can be downloaded from http://dev.opencascade.org separately from OCCT code.)
49For this it is recommended to add a file *DrawAppliInit* in the directory which is current at the moment of starting DRAWEXE (normally it is OCCT root directory, <i>$CASROOT </i>). This file is evaluated automatically at the DRAW start.
51Example (Windows)
72b7576f 52
504a8968 54set env(CSF_TestDataPath) $env(CSF_TestDataPath)\;d:/occt/test-data
72b7576f 55return ;# this is to avoid an echo of the last command above in cout
504a8968 56
72b7576f 57~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
504a8968 58Note that variable *CSF_TestDataPath* is set to default value at DRAW start, pointing at the folder <i>$CASROOT/data</i>.
59In this example, subdirectory <i>d:/occt/test-data</i> is added to this path. Similar code could be used on Linux and Mac OS X except that on non-Windows platforms colon ‘:’ should be used as path separator instead of semicolon ‘;’.
72b7576f 60
504a8968 61All tests are run from DRAW command prompt (run *draw.tcl* or *draw.sh *to start it).
72b7576f 62
63@subsubsection testmanual_1_3_2 Running Tests
504a8968 65To run all tests, type command *testgrid*
72b7576f 66
504a8968 69~~~~~
70Draw[]> testgrid
72b7576f 72
504a8968 73For running only a group or a grid of tests, give additional arguments indicating followed by path to the group and (if needed) the grid name.
72b7576f 74
504a8968 75Example:
72b7576f 76
504a8968 77~~~~~
78Draw[]> testgrid blend simple
72b7576f 80
72b7576f 81
82As the tests progress, the result of each test case is reported.
504a8968 83At the end of the log a summary of test cases is output,
84including the list of detected regressions and improvements, if any.
72b7576f 85
504a8968 87Example:
72b7576f 89~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.tcl}
90 Tests summary
92 CASE 3rdparty export A1: OK
93 ...
94 CASE pipe standard B1: BAD (known problem)
95 CASE pipe standard C1: OK
96 No regressions
97 Total cases: 208 BAD, 31 SKIPPED, 3 IMPROVEMENT, 1791 OK
98 Elapsed time: 1 Hours 14 Minutes 33.7384512019 Seconds
99 Detailed logs are saved in D:/occt/results_2012-06-04T0919
504a8968 102The tests are considered as non-regressive if only OK, BAD (i.e. known problem), and SKIPPED (i.e. not executed, typically because of lack of a data file) statuses are reported. See <a href="#testmanual_3_5">Interpretation of test results</a> for details.
72b7576f 103
504a8968 104The results and detailed logs of the tests are saved by default to a subdirectory of the current folder, whose name is generated automatically using the current date and time, prefixed by word <i>"results_"</i> and Git branch name (if Git is available and current sources are managed by Git).
105If necessary, a non-default output directory can be specified using option <i> –outdir</i> followed by a path to the directory. This directory should be new or empty; use option –overwrite to allow writing results in existing non-empty directory.
72b7576f 106
504a8968 107Example:
109Draw[]> testgrid –outdir d:/occt/last_results -overwrite
111In the output directory, a cumulative HTML report summary.html provides links to reports on each test case. An additional report in JUnit-style XML format can be output for use in Jenkins or other continuous integration system.
113Type <i>help testgrid</i> in DRAW prompt to get help on options supported by *testgrid* command.
115For example:
72b7576f 116
504a8968 117~~~~~
118Draw[3]> help testgrid
119testgrid: Run all tests, or specified group, or one grid
120 Use: testgrid [group [grid]] [options...]
121 Allowed options are:
122 -parallel N: run N parallel processes (default is number of CPUs, 0 to disable)
123 -refresh N: save summary logs every N seconds (default 60, minimal 1, 0 to disable)
124 -outdir dirname: set log directory (should be empty or non-existing)
125 -overwrite: force writing logs in existing non-empty directory
126 -xml filename: write XML report for Jenkins (in JUnit-like format)
72b7576f 128
504a8968 129@subsubsection testmanual_1_3_3 Running a Single Test
131To run a single test, type command *test*’ followed by names of group, grid, and test case.
72b7576f 132
136 Draw[1]> test blend simple A1
137 CASE blend simple A1: OK
138 Draw[2]>
504a8968 141Note that normally an intermediate output of the script is not shown. The detailed log of the test can be obtained after the test execution by running command <i>"dlog get"</i>.
143To see intermediate commands and their output during the test execution, add one more argument
144<i>"echo"</i> at the end of the command line. Note that with this option the log is not collected and summary is not produced.
146@subsubsection testmanual_1_3_4 Creating a New Test
148The detailed rules of creation of new tests are given in <a href="#testmanual_3">section 3</a>. The following short description covers the most typical situations:
150Use prefix “bug” followed by Mantis issue ID and, if necessary, additional suffixes, for naming the test script and DRAW commands specific for this test case.
1521. If the test requires C++ code, add it as new DRAW command(s) in one of files in *QABugs* package. Note that this package defines macros *QVERIFY* and *QCOMPARE*, thus code created for QTest or GoogleTest frameworks can be used with minimal modifications.
1532. Add script(s) for the test case in grid (subfolder) corresponding to the relevant OCCT module of the group bugs <i>($CASROOT/tests/bugs)</i>. See <a href="#testmanual_5_2">the correspondence map</a>.
1543. In the test script:
155 * Load all necessary DRAW modules by command *pload*.
156 * Use command *locate_data_file* to get a path to data files used by test script. (Make sure to have this command not inside catch statement if it is used.)
157 * Use DRAW commands to reproduce the situation being tested.
158 * If test case is added to describe existing problem and the fix is not available, add TODO message for each error to mark it as known problem. The TODO statements must be specific so as to match the actually generated messages but not all similar errors.
159 * Make sure that in case of failure the test produces message containing word “Error” or other recognized by test system as error (see files parse.rules).
1604. If the test case uses data file(s) not yet present in the test database, these can be put to subfolder data of the test grid, and integrated to Git along with the test case.
1615. Check that the test case runs as expected (test for fix: OK with the fix, FAILED without the fix; test for existing problem: BAD), and integrate to Git branch created for the issue.
165* Added files:
167git status –short
168A tests/bugs/heal/data/OCC210a.brep
169A tests/bugs/heal/data/OCC210a.brep
170A tests/bugs/heal/bug210_1
171A tests/bugs/heal/bug210_2
174* Test script
177puts "OCC210 (case 1): Improve FixShape for touching wires"
179restore [locate_data_file OCC210a.brep] a
181fixshape result a 0.01 0.01
182checkshape result
72b7576f 184
185@section testmanual_2 Organization of Test Scripts
187@subsection testmanual_2_1 General Layout
189Standard OCCT tests are located in subdirectory tests of the OCCT root folder ($CASROOT).
72b7576f 190
504a8968 191Additional test folders can be added to the test system by defining environment variable *CSF_TestScriptsPath*. This should be list of paths separated by semicolons (*;*) on Windows
192or colons (*:*) on Linux or Mac. Upon DRAW launch, path to tests sub-folder of OCCT is added at the end of this variable automatically.
194Each test folder is expected to contain:
72b7576f 195 * Optional file parse.rules defining patterns for interpretation of test results, common for all groups in this folder
196 * One or several test group directories.
198Each group directory contains:
504a8968 200 * File *grids.list* that identifies this test group and defines list of test grids in it.
201 * Test grids (sub-directories), each containing set of scripts for test cases, and optional files *cases.list*, *parse.rules*, *begin* and *end*.
72b7576f 202 * Optional sub-directory data
72b7576f 203
504a8968 204By convention, names of test groups, grids, and cases should contain no spaces and be lower-case.
205The names *begin, end, data, parse.rules, grids.list* and *cases.list* are reserved.
72b7576f 206
504a8968 207General layout of test scripts is shown in Figure 1.
209@image html /dev_guides/tests/images/tests_image001.png "Layout of tests folder"
210@image latex /dev_guides/tests/images/tests_image001.png "Layout of tests folder"
72b7576f 211
72b7576f 212
213@subsection testmanual_2_2 Test Groups
215@subsubsection testmanual_2_2_1 Group Names
504a8968 217The names of directories of test groups containing systematic test grids correspond to the functionality tested by each group.
72b7576f 219Example:
504a8968 221~~~~~
72b7576f 222 caf
223 mesh
224 offset
504a8968 225~~~~~
72b7576f 226
504a8968 227Test group *bugs* is used to collect the tests coming from bug reports. Group *demo* collects tests of the test system, DRAW, samples, etc.
72b7576f 228
504a8968 229@subsubsection testmanual_2_2_2 File "grids.list"
231This test group contains file *grids.list*, which defines an ordered list of grids in this group in the following format:
72b7576f 232
234001 gridname1
235002 gridname2
237NNN gridnameN
243 001 basic
244 002 advanced
504a8968 247@subsubsection testmanual_2_2_3 File "begin"
72b7576f 248
504a8968 249This file is a Tcl script. It is executed before every test in the current group.
72b7576f 250Usually it loads necessary Draw commands, sets common parameters and defines
251additional Tcl functions used in test scripts.
504a8968 252
72b7576f 253Example:
256 pload TOPTEST ;# load topological command
257 set cpulimit 300 ;# set maximum time allowed for script execution
504a8968 260@subsubsection testmanual_2_2_4 File "end"
72b7576f 261
504a8968 262This file is a TCL script. It is executed after every test in the current group. Usually it checks the results of script work, makes a snap-shot of the viewer and writes *TEST COMPLETED* to the output.
72b7576f 263
504a8968 264Note: *TEST COMPLETED* string should be present in the output to indicate that the test is finished without crash.
266See <a href="#testmanual_3">section 3</a> for more information.
72b7576f 270 if { [isdraw result] } {
271 checkshape result
272 } else {
273 puts *Error: The result shape can not be built*
274 }
504a8968 276~~~~~
278@subsubsection testmanual_2_2_5 File "parse.rules"
280The test group may contain *parse.rules* file. This file defines patterns used for analysis of the test execution log and deciding the status of the test run.
282Each line in the file should specify a status (single word), followed by a regular expression delimited by slashes (*/*) that will be matched against lines in the test output log to check if it corresponds to this status.
284The regular expressions support a subset of the Perl *re* syntax. See also <a href=http://perldoc.perl.org/perlre.html>Perl regular expressions</a>.
286The rest of the line can contain a comment message, which will be added to the test report when this status is detected.
72b7576f 287
72b7576f 288Example:
504a8968 290~~~~~
e5bd0d98 291 FAILED /\\b[Ee]xception\\b/ exception
292 FAILED /\\bError\\b/ error
72b7576f 293 SKIPPED /Cannot open file for reading/ data file is missing
294 SKIPPED /Could not read file .*, abandon/ data file is missing
504a8968 295~~~~~
72b7576f 296
297Lines starting with a *#* character and blank lines are ignored to allow comments and spacing.
72b7576f 298
504a8968 299See <a href="#testmanual_3_5">Interpretation of test results</a> chapter for details.
301If a line matches several rules, the first one applies. Rules defined in the grid are checked first, then rules in the group, then rules in the test root directory. This allows defining some rules on the grid level with status *IGNORE* to ignore messages that would otherwise be treated as errors due to the group level rules.
72b7576f 303Example:
e5bd0d98 306 FAILED /\\bFaulty\\b/ bad shape
72b7576f 307 IGNORE /^Error [23]d = [\d.-]+/ debug output of blend command
308 IGNORE /^Tcl Exception: tolerance ang : [\d.-]+/ blend failure
504a8968 311@subsubsection testmanual_2_2_6 Directory "data"
312The test group may contain subdirectory *data*, where test scripts shared by different test grids can be put. See also <a href="#testmanual_2_3_5">Directory *data*</a>.
72b7576f 314@subsection testmanual_2_3 Test Grids
316@subsubsection testmanual_2_3_1 Grid Names
504a8968 318The folder of a test group can have several sub-directories (Grid 1… Grid N) defining test grids.
319Each directory contains a set of related test cases. The name of a directory should correspond to its contents.
72b7576f 320
504a8968 323~~~~~
72b7576f 324caf
325 basic
326 bugs
327 presentation
504a8968 328~~~~~
72b7576f 329
504a8968 330Here *caf* is the name of the test group and *basic*, *bugs*, *presentation*, etc. are the names of grids.
72b7576f 331
504a8968 332@subsubsection testmanual_2_3_2 File "begin"
334This file is a TCL script executed before every test in the current grid.
72b7576f 335
72b7576f 336Usually it sets variables specific for the current grid.
504a8968 337
72b7576f 338Example:
341 set command bopfuse ;# command tested in this grid
504a8968 344@subsubsection testmanual_2_3_3 File "end"
346This file is a TCL script executed after every test in current grid.
348Usually it executes a specific sequence of commands common for all tests in the grid.
72b7576f 349
72b7576f 350Example:
353 vdump $logdir/${casename}.gif ;# makes a snap-shot of AIS viewer
504a8968 356@subsubsection testmanual_2_3_4 File "cases.list"
72b7576f 357
358The grid directory can contain an optional file cases.list
504a8968 359defining an alternative location of the test cases.
360This file should contain a single line defining the relative path to the collection of test cases.
72b7576f 361
504a8968 364~~~~~
72b7576f 365../data/simple
504a8968 366~~~~~
368This option is used for creation of several grids of tests with the same data files and operations but performed with differing parameters. The common scripts are usually located place in the common
369subdirectory of the test group, <i>data/simple</i> for example.
72b7576f 370
504a8968 371If file *cases.list* exists, the grid directory should not contain any test cases.
72b7576f 372The specific parameters and pre- and post-processing commands
504a8968 373for test execution in this grid should be defined in the files *begin* and *end*.
376@subsubsection testmanual_2_3_5 Directory "data"
378The test grid may contain subdirectory *data*, containing data files used in tests (BREP, IGES, STEP, etc.) of this grid.
72b7576f 379
380@subsection testmanual_2_4 Test Cases
504a8968 382The test case is a TCL script, which performs some operations using DRAW commands
383and produces meaningful messages that can be used to check the validity of the result.
72b7576f 385Example:
388 pcylinder c1 10 20 ;# create first cylinder
389 pcylinder c2 5 20 ;# create second cylinder
390 ttranslate c2 5 0 10 ;# translate second cylinder to x,y,z
391 bsection result c1 c2 ;# create a section of two cylinders
392 checksection result ;# will output error message if result is bad
504a8968 395The test case can have any name (except for the reserved names *begin, end, data, cases.list* and *parse.rules*).
72b7576f 396For systematic grids it is usually a capital English letter followed by a number.
504a8968 400~~~~~
72b7576f 401 A1
402 A2
403 B1
404 B2
504a8968 405~~~~~
72b7576f 406
504a8968 407Such naming facilitates compact representation of tests execution results in tabular format within HTML reports.
72b7576f 408
72b7576f 409
410@section testmanual_3 Creation And Modification Of Tests
412This section describes how to add new tests and update existing ones.
414@subsection testmanual_3_1 Choosing Group, Grid, and Test Case Name
504a8968 416The new tests are usually added in the frame of processing issues in OCCT Mantis tracker.
72b7576f 417Such tests in general should be added to group bugs, in the grid
504a8968 418corresponding to the affected OCCT functionality. See <a href="#testmanual_5_2">Mapping of OCCT functionality to grid names in group *bugs*</a>.
420New grids can be added as necessary to contain tests for the functionality not yet covered by existing test grids.
421The test case name in the bugs group should be prefixed by the ID of the corresponding issue in Mantis (without leading zeroes) with prefix *bug*. It is recommended to add a suffix providing a hint on the tested situation. If more than one test is added for a bug, they should be distinguished by suffixes; either meaningful or just ordinal numbers.
72b7576f 422
504a8968 426 bug12345_coaxial
427 bug12345_orthogonal_1
428 bug12345_orthogonal_2
72b7576f 429~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
504a8968 431If the new test corresponds to a functionality already covered by the existing systematic test grid (e.g. group *mesh* for *BRepMesh* issues), this test can be added (or moved later by OCC team) to that grid.
72b7576f 432
433@subsection testmanual_3_2 Adding Data Files Required for a Test
504a8968 435It is advisable to make self-contained test scripts whenever possible, so as they could be used in environments where data files are not available. For that simple geometric objects and shapes can be created using DRAW commands in the test script itself.
437If the test requires a data file, it should be put to subdirectory *data* of the test grid. It is recommended to prefix the data file with the corresponding issue id prefixed by *bug*, e.g. *bug12345_face1.brep*, to avoid possible conflicts with names of existing data files.
439Note that when the test is integrated to the master branch, OCC team will move the data file to data files repository, so as to keep OCCT sources repository clean from data files.
441When preparing a test script, try to minimize the size of involved data model. For instance, if the problem detected on a big shape can be reproduced on a single face extracted from that shape, use only that face in the test.
444@subsection testmanual_3_3 Adding new DRAW commands
446If the test cannot be implemented using available DRAW commands, consider the following possibilities:
447* If the existing DRAW command can be extended to enable possibility required for a test in a natural way (e.g. by adding an option to activate a specific mode of the algorithm), this way is recommended. This change should be appropriately documented in a relevant Mantis issue.
448* If the new command is needed to access OCCT functionality not exposed to DRAW previously, and this command can be potentially reused (for other tests), it should be added to the package where similar commands are implemented (use *getsource* DRAW command to get the package name). The name and arguments of the new command should be chosen to keep similarity with the existing commands. This change should be documented in a relevant Mantis issue.
449* Otherwise the new command implementing the actions needed for this particular test should be added in *QABugs* package. The command name should be formed by the Mantis issue ID prefixed by *bug*, e.g. *bug12345*.
451Note that a DRAW command is expected to return 0 in case of a normal completion, and 1 (Tcl exception) if it is incorrectly used (e.g. a wrong number of input arguments). Thus if the new command needs to report a test error, this should be done by outputting an appropriate error message rather than by returning a non-zero value.
454@subsection testmanual_3_4 Script Implementation
456The test should run commands necessary to perform the tested operations, in general assuming a clean DRAW session. The required DRAW modules should be loaded by *pload* command, if it is not done by *begin* script. The messages produced by commands in a standard output should include identifiable messages on the discovered problems if any.
458Usually the script represents a set of commands that a person would run interactively to perform the operation and see its results, with additional comments to explain what happens.
72b7576f 460Example:
504a8968 461~~~~~
462# Simple test of fusing box and sphere
463box b 10 10 10
464sphere s 5
465bfuse result b s
466checkshape result
72b7576f 468
504a8968 469Make sure that file *parse.rules* in the grid or group directory contains a regular expression to catch possible messages indicating the failure of the test.
471For instance, for catching errors reported by *checkshape* command relevant grids define a rule to recognize its report by the word *Faulty*:
474FAILED /\bFaulty\b/ bad shape
477For the messages generated in the script it is recommended to use the word 'Error' in the error message.
72b7576f 478
72b7576f 479Example:
504a8968 481~~~~~
482set expected_length 11
483if { [expr $actual_length - $expected_length] > 0.001 } {
484 puts "Error: The length of the edge should be $expected_length"
488At the end, the test script should output *TEST COMPLETED* string to mark a successful completion of the script. This is often done by the *end* script in the grid.
490When the test script requires a data file, use Tcl procedure *locate_data_file* to get a path to it, instead of putting the path explicitly. This will allow easy move of the data file from OCCT sources repository to the data files repository without the need to update the test script.
72b7576f 491
72b7576f 492Example:
504a8968 494~~~~~
495stepread [locate_data_file CAROSKI_COUPELLE.step] a *
498When the test needs to produce some snapshots or other artefacts, use Tcl variable *logdir* as the location where such files should be put. Command *testgrid* sets this variable to the subdirectory of the results folder corresponding to the grid. Command *test* sets it to <i>$CASROOT/tmp</i> unless it is already defined. Use Tcl variable *casename* to prefix all files produced by the test. This variable is set to the name of the test case.
72b7576f 499
72b7576f 500Example:
504a8968 501~~~~~
502xwd $logdir/${casename}.png
503vdisplay result; vfit
504vdump $logdir/${casename}-axo.png
505vfront; vfit
506vdump $logdir/${casename}-front.png
72b7576f 508
504a8968 509would produce:
72b7576f 515
504a8968 516Note that OCCT must be built with FreeImage support to be able to produce usable images.
72b7576f 517
504a8968 518In order to ensure that the test works as expected in different environments, observe the following additional rules:
519* Avoid using external commands such as *grep, rm,* etc., as these commands can be absent on another system (e.g. on Windows); use facilities provided by Tcl instead.
520* Do not put call to *locate_data_file* in catch statement – this can prevent correct interpretation of the missing data file by the test system.
72b7576f 521
72b7576f 522
504a8968 523@subsection testmanual_3_5 Interpretation of test results
72b7576f 524
504a8968 525The result of the test is evaluated by checking its output against patterns defined in the files *parse.rules* of the grid and group.
72b7576f 526
504a8968 527The OCCT test system recognizes five statuses of the test execution:
528* SKIPPED: reported if a line matching SKIPPED pattern is found (prior to any FAILED pattern). This indicates that the test cannot be run in the current environment; the most typical case is the absence of the required data file.
529* FAILED: reported if a line matching pattern with status FAILED is found (unless it is masked by the preceding IGNORE pattern or a TODO statement), or if message TEST COMPLETED is not found at the end. This indicates that the test has produced a bad or unexpected result, and usually means a regression.
530* BAD: reported if the test script output contains one or several TODO statements and the corresponding number of matching lines in the log. This indicates a known problem . The lines matching TODO statements are not checked against other patterns and thus will not cause a FAILED status.
531* IMPROVEMENT: reported if the test script output contains a TODO statement for which no corresponding line is found. This is a possible indication of improvement (a known problem has disappeared).
532* OK: reported if none of the above statuses have been assigned. This means that the test has passed without problems.
72b7576f 533
504a8968 534Other statuses can be specified in *parse.rules* files, these will be classified as FAILED.
72b7576f 535
504a8968 536For integration of the change to OCCT repository, all tests should return either OK or BAD status.
537The new test created for an unsolved problem should return BAD. The new test created for a fixed problem should return FAILED without the fix, and OK with the fix.
72b7576f 538
504a8968 539@subsection testmanual_3_6 Marking BAD cases
72b7576f 540
504a8968 541If the test produces an invalid result at a certain moment then corresponding bug should be created in the OCCT issue tracker located at http://tracker.dev.opencascade.org, and the problem should be marked as TODO in the test script.
72b7576f 542
504a8968 543The following statement should be added to such a test script:
545puts "TODO BugNumber ListOfPlatforms: RegularExpression"
549* *BugNumber* is the bug ID in the tracker. For example: #12345.
550* *ListOfPlatforms* is a list of platforms at which the bug is reproduced (e.g. Mandriva2008, Windows or All). Note that the platform name is custom for the OCCT test system; it corresponds to the value of environment variable *os_type* defined in DRAW.
72b7576f 551
504a8968 553~~~~~
554Draw[2]> puts $env(os_type)
72b7576f 557
504a8968 558* RegularExpression is a regular expression which should be matched against the line indicating the problem in the script output.
72b7576f 559
72b7576f 560Example:
504a8968 561~~~~~
562puts "TODO #22622 Mandriva2008: Abort .* an exception was raised"
72b7576f 564
504a8968 565The parser checks the test output and if an output line matches the *RegularExpression* then it will be assigned a BAD status instead of FAILED.
72b7576f 566
504a8968 567A separate TODO line must be added for each output line matching an error expression to mark the test as BAD. If not all TODO messages are found in the test log, the test will be considered as possible improvement.
569To mark the test as BAD for an incomplete case (when the final *TEST COMPLETE* message is missing) the expression *TEST INCOMPLETE* should be used instead of the regular expression.
72b7576f 570
504a8968 573~~~~~
574puts "TODO OCC22817 All: exception.+There are no suitable edges"
575puts "TODO OCC22817 All: \\*\\* Exception \\*\\*"
576puts "TODO OCC22817 All: TEST INCOMPLETE"
72b7576f 578
504a8968 579
581@section testmanual_4 Advanced Use
72b7576f 582
583@subsection testmanual_4_1 Running Tests on Older Versions of OCCT
504a8968 585Sometimes it might be necessary to run tests on previous versions of OCCT (<= 6.5.4) that do not include this test system. This can be done by adding DRAW configuration file *DrawAppliInit* in the directory which is current by the moment of DRAW start-up, to load test commands and to define necessary environment.
72b7576f 586
504a8968 587Note: in OCCT 6.5.3, file *DrawAppliInit* already exists in <i>$CASROOT/src/DrawResources</i>, new commands should be added to this file instead of a new one in the current directory.
589For example, let us assume that *d:/occt* contains an up-to-date version of OCCT sources with tests, and the test data archive is unpacked to *d:/test-data*):
592set env(CASROOT) d:/occt
593set env(CSF_TestScriptsPath) $env(CASROOT)/tests
594source $env(CASROOT)/src/DrawResources/TestCommands.tcl
595set env(CSF_TestDataPath) $env(CASROOT)/data;d:/test-data
599Note that on older versions of OCCT the tests are run in compatibility mode and not all output of the test command can be captured; this can lead to absence of some error messages (can be reported as either a failure or an improvement).
601@subsection testmanual_4_2 Adding custom tests
603You can extend the test system by adding your own tests. For that it is necessary to add paths to the directory where these tests are located, and one or more additional data directories, to the environment variables *CSF_TestScriptsPath* and *CSF_TestDataPath*. The recommended way for doing this is using DRAW configuration file *DrawAppliInit* located in the directory which is current by the moment of DRAW start-up.
605Use Tcl command *_path_separator* to insert a platform-dependent separator to the path list.
607For example:
609set env(CSF_TestScriptsPath) \
610 $env(TestScriptsPath)[_path_separator]d:/MyOCCTProject/tests
611set env(CSF_TestDataPath) \
612 d:/occt/test-data[_path_separator]d:/MyOCCTProject/data
613return ;# this is to avoid an echo of the last command above in cout
616@subsection testmanual_4_3 Parallel execution of tests
618For better efficiency, on computers with multiple CPUs the tests can be run in parallel mode. This is default behavior for command *testgrid* : the tests are executed in parallel processes (their number is equal to the number of CPUs available on the system). In order to change this behavior, use option parallel followed by the number of processes to be used (1 or 0 to run sequentially).
72b7576f 619
504a8968 620Note that the parallel execution is only possible if Tcl extension package *Thread* is installed. It is included in *ActiveTcl* package, but can be absent in some Linux distributions. If this package is not available, *testgrid* command will output a warning message.
72b7576f 621
504a8968 622@subsection testmanual_4_4 Checking non-regression of performance, memory, and visualization
72b7576f 623
504a8968 624Some test results are very dependent on the characteristics of the workstation, where they are performed, and thus cannot be checked by comparison with some predefined values. These results can be checked for non-regression (after a change in OCCT code) by comparing them with the results produced by the version without this change. The most typical case is comparing the result obtained in a branch created for integration of a fix (CR***) with the results obtained on the master branch before that change is made.
626OCCT test system provides a dedicated command *testdiff* for comparing CPU time of execution, memory usage, and images produced by the tests.
629testdiff dir1 dir2 [groupname [gridname]] [options...]
631Here *dir1* and *dir2* are directories containing logs of two test runs.
633Possible options are:
634* <i>-save <filename> </i> - saves the resulting log in a specified file (<i>$dir1/diff-$dir2.log</i> by default). HTML log is saved with the same name and extension .html;
635* <i>-status {same|ok|all}</i> - allows filtering compared cases by their status:
636 * *same* - only cases with same status are compared (default);
637 * *ok* - only cases with OK status in both logs are compared;
638 * *all* - results are compared regardless of status;
639* <i>-verbose <level> </i> - defines the scope of output data:
640 * 1 - outputs only differences;
641 * 2 - additionally outputs the list of logs and directories present in one of directories only;
642 * 3 - (by default) additionally outputs progress messages;
72b7576f 643
72b7576f 644Example:
504a8968 645
647Draw[]> testdiff results-CR12345-2012-10-10T08:00 results-master-2012-10-09T21:20
650@section testmanual_5 APPENDIX
652@subsection testmanual_5_1 Test groups
654@subsubsection testmanual_5_1_1 3rdparty
656This group allows testing the interaction of OCCT and 3rdparty products.
660| Grid | Commands | Functionality |
661| :---- | :----- | :------- |
662| export | vexport | export of images to different formats |
663| fonts | vtrihedron, vcolorscale, vdrawtext | display of fonts |
666@subsubsection testmanual_5_1_2 blend
668This group allows testing blends (fillets) and related operations.
670DRAW module: MODELING.
672| Grid | Commands | Functionality |
673| :---- | :----- | :------- |
674| simple | blend | fillets on simple shapes |
675| complex | blend | fillets on complex shapes, non-trivial geometry |
676| tolblend_simple | tolblend, blend | |
677| buildevol | buildevol | |
678| tolblend_buildvol | tolblend, buildevol | use of additional command tolblend |
679| bfuseblend | bfuseblend | |
680| encoderegularity | encoderegularity | |
682@subsubsection testmanual_5_1_3 boolean
684This group allows testing Boolean operations.
686DRAW module: MODELING (packages *BOPTest* and *BRepTest*).
688Grids names are based on name of the command used, with suffixes:
689* <i>_2d</i> – for tests operating with 2d objects (wires, wires, 3d objects, etc.);
690* <i>_simple</i> – for tests operating on simple shapes (boxes, cylinders, toruses, etc.);
691* <i>_complex</i> – for tests dealing with complex shapes.
693| Grid | Commands | Functionality |
694| :---- | :----- | :------- |
695| bcommon_2d | bcommon | Common operation (old algorithm), 2d |
696| bcommon_complex | bcommon | Common operation (old algorithm), complex shapes |
697| bcommon_simple | bcommon | Common operation (old algorithm), simple shapes |
698| bcut_2d | bcut | Cut operation (old algorithm), 2d |
699| bcut_complex | bcut | Cut operation (old algorithm), complex shapes |
700| bcut_simple | bcut | Cut operation (old algorithm), simple shapes |
701| bcutblend | bcutblend | |
702| bfuse_2d | bfuse | Fuse operation (old algorithm), 2d |
703| bfuse_complex | bfuse | Fuse operation (old algorithm), complex shapes |
704| bfuse_simple | bfuse | Fuse operation (old algorithm), simple shapes |
705| bopcommon_2d | bopcommon | Common operation, 2d |
706| bopcommon_complex | bopcommon | Common operation, complex shapes |
707| bopcommon_simple | bopcommon | Common operation, simple shapes |
708| bopcut_2d | bopcut | Cut operation, 2d |
709| bopcut_complex | bopcut | Cut operation, complex shapes |
710| bopcut_simple | bopcut | Cut operation, simple shapes |
711| bopfuse_2d | bopfuse | Fuse operation, 2d |
712| bopfuse_complex | bopfuse | Fuse operation, complex shapes |
713| bopfuse_simple | bopfuse | Fuse operation, simple shapes |
714| bopsection | bopsection | Section |
715| boptuc_2d | boptuc | |
716| boptuc_complex | boptuc | |
717| boptuc_simple | boptuc | |
718| bsection | bsection | Section (old algorithm) |
720@subsubsection testmanual_5_1_4 bugs
722This group allows testing cases coming from Mantis issues.
724The grids are organized following OCCT module and category set for the issue in the Mantis tracker.
725See <a href="#testmanual_5_2">Mapping of OCCT functionality to grid names in group *bugs*</a> for details.
727@subsubsection testmanual_5_1_5 caf
729This group allows testing OCAF functionality.
733| Grid | Commands | Functionality |
734| :---- | :----- | :------- |
735| basic | | Basic attributes |
736| bugs | | Saving and restoring of document |
737| driver | | OCAF drivers |
738| named_shape | | *TNaming_NamedShape* attribute |
739| presentation | | *AISPresentation* attributes |
740| tree | | Tree construction attributes |
741| xlink | | XLink attributes |
743@subsubsection testmanual_5_1_6 chamfer
745This group allows testing chamfer operations.
747DRAW module: MODELING.
749The test grid name is constructed depending on the type of the tested chamfers. Additional suffix <i>_complex</i> is used for test cases involving complex geometry (e.g. intersections of edges forming a chamfer); suffix <i>_sequence</i> is used for grids where chamfers are computed sequentially.
751| Grid | Commands | Functionality |
752| :---- | :----- | :------- |
753| equal_dist | | Equal distances from edge |
754| equal_dist_complex | | Equal distances from edge, complex shapes |
755| equal_dist_sequence | | Equal distances from edge, sequential operations |
756| dist_dist | | Two distances from edge |
757| dist_dist_complex | | Two distances from edge, complex shapes |
758| dist_dist_sequence | | Two distances from edge, sequential operations |
759| dist_angle | | Distance from edge and given angle |
760| dist_angle_complex | | Distance from edge and given angle |
761| dist_angle_sequence | | Distance from edge and given angle |
763@subsubsection testmanual_5_1_7 demo
765This group allows demonstrating how testing cases are created, and testing DRAW commands and the test system as a whole.
767| Grid | Commands | Functionality |
768| :---- | :----- | :------- |
769| draw | getsource, restore | Basic DRAW commands |
770| testsystem | | Testing system |
771| samples | | OCCT samples |
774@subsubsection testmanual_5_1_8 draft
776This group allows testing draft operations.
778DRAW module: MODELING.
780| Grid | Commands | Functionality |
781| :---- | :----- | :------- |
782| Angle | depouille | Drafts with angle (inclined walls) |
785@subsubsection testmanual_5_1_9 feat
787This group allows testing creation of features on a shape.
789DRAW module: MODELING (package *BRepTest*).
791| Grid | Commands | Functionality |
792| :---- | :----- | :------- |
793| featdprism | | |
794| featlf | | |
795| featprism | | |
796| featrevol | | |
797| featrf | | |
799@subsubsection testmanual_5_1_10 heal
801This group allows testing the functionality provided by *ShapeHealing* toolkit.
803DRAW module: XSDRAW
805| Grid | Commands | Functionality |
806| :---- | :----- | :------- |
807| fix_shape | fixshape | Shape healing |
808| fix_gaps | fixwgaps | Fixing gaps between edges on a wire |
809| same_parameter | sameparameter | Fixing non-sameparameter edges |
810| fix_face_size | DT_ApplySeq | Removal of small faces |
811| elementary_to_revolution | DT_ApplySeq | Conversion of elementary surfaces to revolution |
812| direct_faces | directfaces | Correction of axis of elementary surfaces |
813| drop_small_edges | fixsmall | Removal of small edges |
814| split_angle | DT_SplitAngle | Splitting periodic surfaces by angle |
815| split_angle_advanced | DT_SplitAngle | Splitting periodic surfaces by angle |
816| split_angle_standard | DT_SplitAngle | Splitting periodic surfaces by angle |
817| split_closed_faces | DT_ClosedSplit | Splitting of closed faces |
818| surface_to_bspline | DT_ToBspl | Conversion of surfaces to b-splines |
819| surface_to_bezier | DT_ShapeConvert | Conversion of surfaces to bezier |
820| split_continuity | DT_ShapeDivide | Split surfaces by continuity criterion |
821| split_continuity_advanced | DT_ShapeDivide | Split surfaces by continuity criterion |
822| split_continuity_standard | DT_ShapeDivide | Split surfaces by continuity criterion |
823| surface_to_revolution_advanced | DT_ShapeConvertRev | Convert elementary surfaces to revolutions, complex cases |
824| surface_to_revolution_standard | DT_ShapeConvertRev | Convert elementary surfaces to revolutions, simple cases |
826@subsubsection testmanual_5_1_11 mesh
828This group allows testing shape tessellation (*BRepMesh)) and shading.
830DRAW modules: MODELING (package *MeshTest*), VISUALIZATION (package *ViewerTest*)
832| Grid | Commands | Functionality |
833| :---- | :----- | :------- |
834| advanced_shading | vdisplay | Shading, complex shapes |
835| standard_shading | vdisplay | Shading, simple shapes |
836| advanced_mesh | mesh | Meshing of complex shapes |
837| standard_mesh | mesh | Meshing of simple shapes |
838| advanced_incmesh | incmesh | Meshing of complex shapes |
839| standard_incmesh | incmesh | Meshing of simple shapes |
840| advanced_incmesh_parallel | incmesh | Meshing of complex shapes, parallel mode |
841| standard_incmesh_parallel | incmesh | Meshing of simple shapes, parallel mode |
843@subsubsection testmanual_5_1_12 mkface
845This group allows testing creation of simple surfaces.
847DRAW module: MODELING (package *BRepTest*)
849| Grid | Commands | Functionality |
850| :---- | :----- | :------- |
851| after_trim | mkface | |
852| after_offset | mkface | |
853| after_extsurf_and_offset | mkface | |
854| after_extsurf_and_trim | mkface | |
855| after_revsurf_and_offset | mkface | |
856| mkplane | mkplane | |
858@subsubsection testmanual_5_1_13 nproject
860This group allows testing normal projection of edges and wires onto a face.
862DRAW module: MODELING (package *BRepTest*)
864| Grid | Commands | Functionality |
865| :---- | :----- | :------- |
866| Base | nproject | |
868@subsubsection testmanual_5_1_14 offset
870This group allows testing offset functionality for curves and surfaces.
872DRAW module: MODELING (package *BRepTest*)
874| Grid | Commands | Functionality |
875| :---- | :----- | :------- |
876| compshape | offsetcompshape | Offset of shapes with removal of some faces |
877| faces_type_a | offsetparameter, offsetload, offsetperform | Offset on a subset of faces with a fillet |
878| faces_type_i | offsetparameter, offsetload, offsetperform | Offset on a subset of faces with a sharp edge |
879| shape_type_a | offsetparameter, offsetload, offsetperform | Offset on a whole shape with a fillet |
880| shape_type_i | offsetparameter, offsetload, offsetperform | Offset on a whole shape with a fillet |
881| shape | offsetshape | |
882| wire_closed_outside_0_005, wire_closed_outside_0_025, wire_closed_outside_0_075, wire_closed_inside_0_005, wire_closed_inside_0_025, wire_closed_inside_0_075, wire_unclosed_outside_0_005, wire_unclosed_outside_0_025, wire_unclosed_outside_0_075 | mkoffset | 2d offset of closed and unclosed planar wires with different offset step and directions of offset ( inside / outside ) |
884@subsubsection testmanual_5_1_15 pipe
886This group allows testing construction of pipes (sweeping of a contour along profile).
888DRAW module: MODELING (package *BRepTest*)
890| Grid | Commands | Functionality |
891| :---- | :----- | :------- |
892| Standard | pipe | |
894@subsubsection testmanual_5_1_16 prism
896This group allows testing construction of prisms.
898DRAW module: MODELING (package *BRepTest*)
900| Grid | Commands | Functionality |
901| :---- | :----- | :------- |
902| seminf | prism | |
904@subsubsection testmanual_5_1_17 sewing
906This group allows testing sewing of faces by connecting edges.
908DRAW module: MODELING (package *BRepTest*)
910| Grid | Commands | Functionality |
911| :---- | :----- | :------- |
912| tol_0_01 | sewing | Sewing faces with tolerance 0.01 |
913| tol_1 | sewing | Sewing faces with tolerance 1 |
914| tol_100 | sewing | Sewing faces with tolerance 100 |
916@subsubsection testmanual_5_1_18 thrusection
918This group allows testing construction of shell or a solid passing through a set of sections in a given sequence (loft).
920| Grid | Commands | Functionality |
921| :---- | :----- | :------- |
922| solids | thrusection | Lofting with resulting solid |
923| not_solids | thrusection | Lofting with resulting shell or face |
925@subsubsection testmanual_5_1_19 xcaf
927This group allows testing extended data exchange packages.
929| Grid | Commands | Functionality |
930| :---- | :----- | :------- |
931| dxc, dxc_add_ACL, dxc_add_CL, igs_to_dxc, igs_add_ACL, brep_to_igs_add_CL, stp_to_dxc, stp_add_ACL, brep_to_stp_add_CL, brep_to_dxc, add_ACL_brep, brep_add_CL | | Subgroups are divided by format of source file, by format of result file and by type of document modification. For example, *brep_to_igs* means that the source shape in brep format was added to the document, which was saved into igs format after that. The postfix *add_CL* means that colors and layers were initialized in the document before saving and the postfix *add_ACL* corresponds to the creation of assembly and initialization of colors and layers in a document before saving. |
934@subsection testmanual_5_2 Mapping of OCCT functionality to grid names in group *bugs*
936| OCCT Module / Mantis category | Toolkits | Test grid in group bugs |
937| :---------- | :--------- | :---------- |
938| Application Framework | PTKernel, TKPShape, TKCDF, TKLCAF, TKCAF, TKBinL, TKXmlL, TKShapeSchema, TKPLCAF, TKBin, TKXml, TKPCAF, FWOSPlugin, TKStdLSchema, TKStdSchema, TKTObj, TKBinTObj, TKXmlTObj | caf |
939| Draw | TKDraw, TKTopTest, TKViewerTest, TKXSDRAW, TKDCAF, TKXDEDRAW, TKTObjDRAW, TKQADraw, DRAWEXE, Problems of testing system | draw |
940| Shape Healing | TKShHealing | heal |
941| Mesh | TKMesh, TKXMesh | mesh |
942| Data Exchange | TKIGES | iges |
943| Data Exchange | TKSTEPBase, TKSTEPAttr, TKSTEP209, TKSTEP | step |
944| Data Exchange | TKSTL, TKVRML | stlvrml |
946| Foundation Classes | TKernel, TKMath, TKAdvTools | fclasses |
947| Modeling_algorithms | TKGeomAlgo, TKTopAlgo, TKPrim, TKBO, TKBool, TKHLR, TKFillet, TKOffset, TKFeat, TKXMesh | modalg |
948| Modeling Data | TKG2d, TKG3d, TKGeomBase, TKBRep | moddata |
949| Visualization | TKService, TKV2d, TKV3d, TKOpenGl, TKMeshVS, TKNIS, TKVoxel | vis |
952@subsection testmanual_5_2 Recommended approaches to checking test results
954@subsubsection testmanual_5_3_1 Shape validity
956Run command *checkshape* on the result (or intermediate) shape and make sure that *parse.rules* of the test grid or group reports bad shapes (usually recognized by word "Faulty") as error.
960checkshape result
963@subsubsection testmanual_5_3_2 Shape tolerance
964The maximal tolerance of sub-shapes of each kind of the resulting shape can be extracted from output of tolerance command as follows:
967set tolerance [tolerance result]
968regexp { *FACE +: +MAX=([-0-9.+eE]+)} $tolerance dummy max_face
969regexp { *EDGE +: +MAX=([-0-9.+eE]+)} $tolerance dummy max_edgee
970regexp { *VERTEX +: +MAX=([-0-9.+eE]+)} $tolerance dummy max_vertex
973@subsubsection testmanual_5_3_3 Shape volume, area, or length
975Use command *vprops, sprops,* or *lprops* to correspondingly measure volume, area, or length of the shape produced by the test. The value can be extracted from the result of the command by *regexp*.
979# check area of shape result with 1% tolerance
980regexp {Mass +: +([-0-9.+eE]+)} [sprops result] dummy area
981if { abs($area - $expected) > 0.1 + 0.01 * abs ($area) } {
982 puts "Error: The area of result shape is $area, while expected $expected"
986@subsubsection testmanual_5_3_4 Memory leaks
988The test system measures the amount of memory used by each test case, and considerable deviations (as well as overall difference) comparing with reference results will be reported by *testdiff* command.
990The typical approach to checking memory leak on a particular operation is to run this operation in cycle measuring memory consumption at each step and comparing it with some threshold value. Note that file begin in group bugs defines command *checktrend* that can be used to analyze a sequence of memory measurements to get statistically based evaluation of the leak presence.
994set listmem {}
995for {set i 1} {$i < 100} {incr i} {
996 # run suspect operation
998 # check memory usage (with tolerance equal to half page size)
999 lappend listmem [expr [meminfo w] / 1024]
1000 if { [checktrend $listmem 0 256 "Memory leak detected"] } {
1001 puts "No memory leak, $i iterations"
1002 break
1003 }
1007@subsubsection testmanual_5_3_5 Visualization
1009Take a snapshot of the viewer, give it the name of the test case, and save in the directory indicated by Tcl variable *imagedir*. Note that this variable directs to the *log* directory if command *testgrid* is active, or to *tmp* subdirectory of the current folder if the test is run interactively.
1014vdisplay result
1015vsetdispmode 1
1018vdump $imagedir/${casename}_shading.png
1021This image will be included in the HTML log produced by *testgrid* command and will be checked for non-regression through comparison of images by command *testdiff*.