ReqIF is a Python library for working with ReqIF format.
- Parsing/unparsing ReqIF
- Formatting (pretty-printing) ReqIF
- Basic validation of ReqIF
- Anonymizing ReqIF files to safely exchange the problematic ReqIF files.
pip install reqiffrom reqif.parser import ReqIFParser
input_file_path = "input.reqif"
reqif_bundle = ReqIFParser.parse(input_file_path)
for specification in reqif_bundle.core_content.req_if_content.specifications:
print(specification.long_name)
for current_hierarchy in reqif_bundle.iterate_specification_hierarchy(specification):
print(current_hierarchy)or for ReqIFz files:
from reqif.parser import ReqIFZParser
input_file_path = "input.reqifz"
reqifz_bundle = ReqIFZParser.parse(input_file_path)
for reqif_bundle_file, reqif_bundle in reqifz_bundle.reqif_bundles.items():
print(f"Bundle: {reqif_bundle_file} {reqif_bundle}")
for specification in reqif_bundle.core_content.req_if_content.specifications:
print(specification)
for attachment in reqifz_bundle.attachments:
print(f"Attachment: {attachment}")from reqif.parser import ReqIFParser
from reqif.unparser import ReqIFUnparser
input_file_path = "input.sdoc"
output_file_path = "output.sdoc"
reqif_bundle = ReqIFParser.parse(input_file_path)
reqif_xml_output = ReqIFUnparser.unparse(reqif_bundle)
with open(output_file_path, "w", encoding="UTF-8") as output_file:
output_file.write(reqif_xml_output)The contents of reqif_xml_output should be the same as the contents of the
input_file.
After installing the reqif Pip package, the reqif command becomes available
and can be executed from the command line.
reqif
usage: reqif [-h] {passthrough,anonymize,dump,format,validate,version} ...
reqif: error: the following arguments are required: command
The validate command is the first command to run against a ReqIF file. The
reqif library contains two sets of checks:
- reqif-internal checks of a ReqIF file's schema and semantics.
- a check of a ReqIF file against the ReqIF's official schema maintained by the Object Management Group (OMG).
The first set of checks is always enabled. To enable the second set of the
official ReqIF schema checks, use --use-reqif-schema:
reqif validate --use-reqif-schema sample.reqif
If an error is found, the reqif validate command exits with 1. If no errors
are found, the exit code is 0.
The reqif passthrough command is a useful tool for verifying whether the reqif
tool correctly parses a given ReqIF file format that the user has at hand.
The passthrough command first parses the ReqIF XML into in-memory Python
objects and then unparses these Python objects back to an output ReqIF file.
If everything goes fine, the output of the passthrough command should be identical to the contents of the input file.
tests/integration/examples contains samples of ReqIF files found on the
internet. The integration tests ensure that for these samples, the passthrough
command always produces outputs that are identical to inputs.
The format command is similar to clang-format for C/C++ files or
cmake-format for CMake files. The input file is parsed and then pretty-printed
back to an output file.
This command is useful when dealing with the ReqIF files that are hand-written or the ReqIF files produced by the ReqIF tools that do not generate well-formed XML with consistent indentation.
The tests/integration/commands/format contains typical examples of
incorrectly formatted ReqIF files. The integration tests ensure that the
format command fixes these issues.
The anonymization helps when exchanging ReqIF documents between different ReqIF
tools including this reqif library. If a particular file is not recognized
correctly by a tool, a user can send their anonymized file to a developer for
further inspection.
The anonymize command accepts an input .reqif file and produces an anonymized
version of that file to the output .reqif file.
usage: reqif anonymize [-h] input_file output_file
main.py anonymize: error: the following arguments are required: input_file, output_file
Examples of anonymization:
...
<ATTRIBUTE-VALUE-STRING THE-VALUE="...Anonymized-2644691225...">
...
<ATTRIBUTE-VALUE-XHTML>
<DEFINITION>
<ATTRIBUTE-DEFINITION-XHTML-REF>rmf-7d0ed062-e964-424c-8305-45067118d959</ATTRIBUTE-DEFINITION-XHTML-REF>
</DEFINITION>
<THE-VALUE>...Anonymized-141441514...</THE-VALUE>
...The anonymization algorithm preserves the uniqueness of the anonymized strings in the document. This way, if the requirement UID identifiers are anonymized, they will still be unique strings in an anonymized document.
The tests/integration/examples folder contains various examples of how the
ReqIF library can be used. Some examples could be and will be made more
advanced over time. The examples are stored in the integration tests folder, so
that they don't regress over time. The feedback and other example requests are
welcome.
The core of the library is a ReqIF first-stage parser that only transforms the contents of a ReqIF XML file into a ReqIF in-memory representation. The in-memory representation is a tree of Python objects that map directly to the objects of the ReqIF XML file structure (e.g, Spec Objects, Spec Types, Data Types, Specifications, etc.).
The first-stage parser (implemented by the class ReqIFParser) can be used by
user's second-stage parser/converter scripts that convert the ReqIF in-memory
structure into a desired format such as Excel, HTML or other formats. The
two-stage process allows the first stage parsing to focus solely on creating an
in-memory ReqIF object tree, while the second stage parsing can further parse
the ReqIF object tree according to the logical structure of user's documents as
encoded in the ReqIF XML file that was produced by user's requirements
management tool.
The reverse process is also possible. A user's script converts another format's
contents into a ReqIF in-memory representation. The ReqIF un-parser
(implemented by the class ReqIFUnparser) can be used to render the in-memory
objects to the ReqIF XML file.
The first-stage parser is made tolerant against possible issues in ReqIF. It should be possible to parse a ReqIF file even if it is missing important information.
A minimum ReqIF parsed by the reqif:
<?xml version="1.0" encoding="UTF-8"?>
<REQ-IF xmlns="http://www.omg.org/spec/ReqIF/20110401/reqif.xsd" xmlns:configuration="http://eclipse.org/rmf/pror/toolextensions/1.0">
</REQ-IF>A separate validation command shall be used to confirm the validity of the ReqIF contents.
The reqif library uses a simple convention for printing the XML attributes: the
attributes are always printed in the alphabetic order of their attribute names.
<DATATYPE-DEFINITION-REAL ACCURACY="10" IDENTIFIER="ID_TC1000_DatatypeDefinitionReal" LAST-CHANGE="2012-04-07T01:51:37.112+02:00" LONG-NAME="TC1000 DatatypeDefinitionReal" MAX="1234.5678" MIN="-1234.5678"/>Some tools do not respect this rule: for example some tools will print
the attribute ACCURACY="10" after the LONG-NAME attribute but the reqif
library does not provide support for preserving a non-alphabetic order of the
attributes.
- ReqIF is a standard. See reference document RD01.
- ReqIF has a fixed structure (see "What is common for all ReqIF documents" below)
- ReqIF standard does not define a document structure for every documents so a ReqIF tool implementor is free to choose between several implementation approaches. There is a ReqIF Implementation Guide that attempts to harmonize ReqIF tool developments. See also "What is left open by the ReqIF standard" below.
- ReqIF files produced by various tool often have incomplete schemas.
- All documents have ReqIF tags:
- Document metadata is stored inside tags of
REQ-IF-HEADERtag. - Requirements are stored as
<SPEC-OBJECT>s - Requirements types are stored as
<SPEC-TYPE>s - Supported data types are stored as
<DATATYPE> - Relationships such as 'Parent-Child' as stored as
<SPEC-RELATIONS>
- Document metadata is stored inside tags of
- How to distinguish requirements from headers/sections?
- One way: create separate
SPEC-TYPES: one or more for requirements and one for sections. - Another way: have one spec type but have it provide a
TYPEfield that can be used to distinguish betweenREQUIREMENTorSECTION. - Yet another way: Check if the "ReqIF.ChapterName" is present on the spec object. When present, it is a section. When not, it is a requirement.
- One way: create separate
The latest version is 1.2: Requirements Interchange Format