ReqIF

ReqIF is a Python library for working with ReqIF format.

Supported features

• Parsing/unparsing ReqIF
• Formatting (pretty-printing) ReqIF
• Basic validation of ReqIF
• Anonymizing ReqIF files to safely exchange the problematic ReqIF files.

Getting started

pip install reqif

Using ReqIF as a library

Parsing ReqIF

from 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}")

Unparsing ReqIF

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.

Using ReqIF as a command-line tool

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

Validate command

The validate command is the first command to run against a ReqIF file. The reqif library contains two sets of checks:

1. reqif-internal checks of a ReqIF file's schema and semantics.
2. 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.

Passthrough command

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.

Formatting ReqIF

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.

Anonymizing ReqIF

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.

How-to examples

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.

Implementation details

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.).

Parsing: Converting from ReqIF to other formats

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.

Unparsing: Converting from other formats to ReqIF

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.

Tolerance

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.

Printing of the attributes

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.

A bottom-up overview of the ReqIF format

• 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.

What is common for all ReqIF documents

• All documents have ReqIF tags:
  • Document metadata is stored inside tags of REQ-IF-HEADER tag.
  • 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>

What is left open by the ReqIF standard

• 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 TYPE field that can be used to distinguish between REQUIREMENT or SECTION.
  • 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.

Reference documents

[RD01] ReqIF standard

The latest version is 1.2: Requirements Interchange Format

[RD02] ReqIF Implementation Guide

ReqIF Implementation Guide v1.8