HRML parser

Description

This sample application parses text in hrml format. The description of the task was suggested by the link. Application can use two approaches: The first with tag tree building, the second with direct stream search. Approaches are switched with --parser option.

Building application

To build application, CMake minimum version 3.10 and compiler with C++17 support required. Use following steps.

• git clone https://github.com/idales/hrml_parserto clone repository.
• mkdir build && cd build in repository folder to create build folder.
• cmake .. to create cmake configuration.
• cmake --build . to build the application.

Command line parameters

Usage: hrml_parser [--parser|-p <parser type>] <input filename>

<parser type> could be "tag_tree" or "in_stream"

• various algorithms for finding attribute value

<input filename> input file with following format

• Lines in HRML format
• Empty line
• Lines with queries in the format <tag>[.<tag>]... ~<attribute>
• Empty line

HRML format

HRML is artifical markup language. Each line has only one opening or closing tag. Opening tags can also contain attributes in form <attribute name> = "<attribute value>". Tags can be nested.

Sample input and output

Input file can be as follows:

        <tag1 value = "Hello World">
        <tag2 name = "Name1">
        </tag2>
        <tag3 another="another" final="final">
        </tag3>
        </tag1>

        tag1.tag2~name
        tag1~name
        tag1~value

Output should be:

        tag1.tag2~name="Name1"
        tag1~name not found
        tag1~value="Hello World"