You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
However. Some people are obsessed with writing everything for themselves. We
Thank you for your work on Boost Parser. It's an impressive library, and I've found it quite powerful.
While I appreciate the humor in the writing, I found the tone to be slightly dismissive of scenarios where writing custom parsers might genuinely be the most effective solution. For instance, I am working on parsing SystemVerilog code, which includes number literals with unique syntax, such as 5 'D 3. Parsing such constructs character by character using char_ primitives seems neither convenient nor effective, and it would make a strong case for writing a custom parser akin to uint_.
I believe this use case illustrates that there are situations where users might reasonably need more guidance on creating low-level parsers. If this section of the documentation could include:
A more neutral tone when addressing the use of custom parsers.
An example of how to effectively handle complex cases (like SystemVerilog number literals) using the library's primitives or via a custom parser.
This approach would make the documentation more inclusive and better serve a wider range of users.
Thank you again for your efforts, and I hope my feedback helps in further refining the excellent resources you provide.
The text was updated successfully, but these errors were encountered:
parser/doc/tutorial.qbk
Line 3792 in f468d52
Thank you for your work on Boost Parser. It's an impressive library, and I've found it quite powerful.
While I appreciate the humor in the writing, I found the tone to be slightly dismissive of scenarios where writing custom parsers might genuinely be the most effective solution. For instance, I am working on parsing SystemVerilog code, which includes number literals with unique syntax, such as
5 'D 3
. Parsing such constructs character by character using char_ primitives seems neither convenient nor effective, and it would make a strong case for writing a custom parser akin touint_
.I believe this use case illustrates that there are situations where users might reasonably need more guidance on creating low-level parsers. If this section of the documentation could include:
This approach would make the documentation more inclusive and better serve a wider range of users.
Thank you again for your efforts, and I hope my feedback helps in further refining the excellent resources you provide.
The text was updated successfully, but these errors were encountered: