Skip to content

Latest commit

 

History

History
138 lines (102 loc) · 5.64 KB

issues.md

File metadata and controls

138 lines (102 loc) · 5.64 KB
Raw

Bugs - General

Priority 1 - Must Solve Before Initial

Priority 2 - Like To Solve Before Initial

  • make sure to generated ordered/unordered tests to make sure both covered
    • every unordered tests should have an ordered counterpart
    • every ordered tests should have an unordered counterpart

nested lists

  • 4+ levels of nesting
    • test_list_blocks_271fx and test_list_blocks_271fa

Like To Have - Issues

  • better solution than no-space-in-code for scenario-cases.md
  • take __consume_text_for_image_alt_text and other functions like it and move as much as possible into token classes
    • TransformToGfmListLooseness.__handle_block_quote_start

Like To Have - Features

  • documentation for developers

  • yaml support for front-matter

    • if front matter is found to be disqualified, send error?

  • better globbing for 043

Priority 3 - Like To Solve in next 3 m

Rule ???

  • is there a rule that can detect a possible front-matter header and trigger? i.e. something looks like front-matter and should be interpretted as front-matter
  • within a given list or block quote, should only have one pattern of container tokens

Other

  • md023 - can line 61 search be done more efficiently by looking for {space}\x02 ?

  • in inline, does if coalesced_stack and coalesced_stack[-1].is_block_quote_start: really need to look for any bq on the stack?

  • show url for failed rules as option

  • combine traversal for 027 and 007?

  • determine which errors to print and which to log

  • (performance) debug: search for any cases where adding a call to is_debug_enabled would be helpful

  • (performance) assert: do as little as possible to ensure better performance

  • (performance) else: ??

  • (performance) replace MarkdownToken properties like is_block_quote_end with more performant?

  • (performance) count_characters_in_text

  • (performance) can further eliminate len() calls by looking for places can calculate once and pass to child functions

  • (performance) order of "if" conditions in critical areas

  • (performance) are all "x and x.blah" requiring the "x" check first? can factor out?

  • (performance) ParserHelper functions can be further optimized? i.e. __remove_backspaces_from_text and how combines

  • (performance) handle_inline_backslash

  • (performance) handle_character_reference

  • (maintenance) look for cases where = True or = False, and see if can do an assignment easlier i.e. if x: d_a_d = False else: d_a_d = True

  • (maintenance) clean up pylint warning suppressions where possible

    • too-many-instance?

  • more tests that include tab characters

  • look for places where common access patterns can be used i.e. link_title

    • = len(parser_state.token_stack

  • HTML and Fenced Blocks and better handling of capturing newlines to avoid counting token height

  • cases like 183 where first non-ws character is an inline specifier

    • can this be fixed in a way that does not require the other fix to add starting_whitespace after the fact?

  • make better use of index_indent

  • why does __revert_token_to_normal_text_token require TextMarkdown.copy instead of replacement_token = text_token_to_replace.create_copy()?

  • image token handling confusing and non-standard

  • fenced code block handling of line/col could be better

  • LRD and block quote interaction - process_link_reference_definition - search for XXXXX

  • LOGGER.debug("handle_block_quote_section>>fenced") and collapse?

Priority 4 - Like To Solve in next 6 m

  • verifying vs validating?
  • bqp:check_for_lazy_handling, why exclude tb?
  • unify 2 separate calculations in __pre_list function
  • raise requeue_line_info instead of returning it? worried about recursion points
  • replace calculate_last_line with calculate_deltas with some work?
  • is_complete_html_end_tag can be more concise, but worth it?
  • modify_end_string. inline?
  • is_valid_tag_name refactor using ParserHelper?
    • extract_html_attribute_name and others to use "in" where possible instead of "or"
  • __calculate_full_deltas - calc_deltas from ParserHelper?
  • __process_inline_text_block - calc_deltas from ParserHelper?
  • is_length_less_than_or_equal_to(extracted_whitespace, 3) - calc once at start of leaf block proc?
  • is_character_at_index_one_of inside of where? collect function?
  • __parse_line_for_leaf_blocks - why calling count_of_block_quotes_on_stack so often?

Bugs - General - Uncategorized

  • None

Ask talk.commonmark.org

  • test_link_reference_definitions_extra_02b vs test_link_reference_definitions_extra_02c

Features - Extensions

  • front matter
  • tables
  • task list items
  • strikethrough
  • disallowed raw html
  • autolinks extension

Features - Correctness

  • go back and see if can replace some of the end token fiddling with start_markdown_token
  • go through any case that uses lazy and do un-lazy example
  • samples that end without a blank line, and add a blank line?

Features - Performance

  • reduce html_helper functions?
    • i.e. html_helper contains 2 for various elements?
  • collect_until_one_of_characters with backslashes?
  • rewrite transform to allow it to consume a Markdown file as it goes
  • modify parse_blocks_pass to consume lines as it goes, instead of requiring entire string in memory

document current restrictions

  • nested block quotes and nested lists are okay, but not together

test_whitespaces_block_quotes_with_tabs_triple_space_space_tab_double test_whitespaces_block_quotes_with_tabs_triple_space_tab_space_double test_whitespaces_block_quotes_with_tabs_triple_space_tab_tab_double

test_whitespaces_block_quotes_with_tabs_triple_space_space_tab_split_double test_whitespaces_block_quotes_with_tabs_triple_space_tab_split_space_double test_whitespaces_block_quotes_with_tabs_triple_space_tab_split_tab_split_double