Commit 84408133 authored by Jan Reimes's avatar Jan Reimes
Browse files

feat(01-normalization-progress-bars): consolidate normalization functions 

- Move normalize_working_group_alias and normalize_subgroup_alias from meetings/utils.py to normalization.py
- Add imports for WorkingGroup and SubWorkingGroup enums to normalization.py
- Update meetings/utils.py to re-export from normalization.py (DRY pattern)
- Functions remain with identical implementation
parent b2fcd324

src/tdoc_crawler/meetings/utils.py

+11 −75
Original line number Diff line number Diff line 
"""Utility functions for meeting data normalization.



Note: normalize_portal_meeting_name is now centralized in tdoc_crawler.utils.normalization.

Import directly from there:



    from tdoc_crawler.utils.normalization import normalize_portal_meeting_name

"""



from __future__ import annotations



import re



from tdoc_crawler.models.subworking_groups import SubWorkingGroup

from tdoc_crawler.models.working_groups import WorkingGroup





def normalize_working_group_alias(alias: str) -> WorkingGroup:

    """Normalize working group aliases to canonical working group enums.



    Supports: RAN, SA, CT and their common name variants.



    Args:

        alias: Working group alias or name



    Returns:

        Canonical working group enum value (WorkingGroup.RAN, WorkingGroup.SA, WorkingGroup.CT)



    Raises:

        ValueError: When the alias cannot be resolved

    """

    alias_upper = alias.strip().upper()

    cleaned = re.sub(r"[#\-_\s]", "", alias_upper)



    # Easy check via list(WorkingGroup): match against name or value

    for working_group in WorkingGroup:

        if cleaned.startswith(working_group.value) or cleaned.startswith(working_group.name):

            return working_group



    # If not found: match first letter

    if cleaned:

        for working_group in WorkingGroup:

            if working_group.value[0] == cleaned[0]:

                return working_group



    # If not found: any match of name/value in cleaned

    for working_group in WorkingGroup:

        if working_group.value in cleaned or working_group.name in cleaned:

            return working_group



    raise ValueError(f"Unknown working group: {alias}")

This module re-exports normalization functions from tdoc_crawler.utils.normalization.

The canonical source for normalization logic is tdoc_crawler.utils.normalization.



Import directly from there when possible:



def normalize_subgroup_alias(alias: str) -> SubWorkingGroup:

    """Normalize subgroup aliases to canonical subgroup enums.



    Returns a SubWorkingGroup enum value (e.g., SubWorkingGroup.S4).

    Supports:

    - Short codes: S4, R1, C3

    - Full names: SA4, RAN1, CT3

    - Separators: SA-4, SA#4, SA 4

    - Plenary: RP, SP, CP, RAN PLENARY, etc.



    Args:

        alias: Subgroup alias or code

    from tdoc_crawler.utils.normalization import normalize_portal_meeting_name



    Returns:

        Canonical subgroup enum value (e.g., SubWorkingGroup.S4)

"""

    input_str = alias.strip().upper()

    cleaned = re.sub(r"[#\-_\s]", "", input_str)



    # determine WorkingGroup first

    wg = normalize_working_group_alias(cleaned)



    # check if cleaned contains a _single_ number or the letter "P"

    match = re.search(r"(P|\d+)", cleaned)

    if not match:

        raise ValueError(f"Cannot parse/normalize subgroup format: {alias}")

    elif len(match.groups()) != 1:

        raise ValueError(f"Multiple/ambiguous matches found when parsing subgroup: {alias}")



    return SubWorkingGroup.from_wg_and_nbr(wg, match.group(1))  # e.g., S4, R1, C3

from __future__ import annotations



from tdoc_crawler.utils.normalization import (

    normalize_portal_meeting_name,

    normalize_subgroup_alias,

    normalize_working_group_alias,

)



__all__ = [

    "normalize_portal_meeting_name",

    "normalize_subgroup_alias",

    "normalize_working_group_alias",

]

src/tdoc_crawler/utils/normalization.py

+71 −0
Original line number Diff line number Diff line
@@ -3,6 +3,9 @@ 
import re

from collections.abc import Generator



from tdoc_crawler.models.subworking_groups import SubWorkingGroup

from tdoc_crawler.models.working_groups import WorkingGroup



_DOTTED_BODY_PATTERN = re.compile(r"^(?P<series>\d{2})\s*\.\s*(?P<increment>\d{1,3})$")

_UNDOTTED_BODY_PATTERN = re.compile(r"^(?P<series>\d{2})(?P<increment>\d{1,3})$")

_OFFSET_PATTERN = re.compile(r"^(?P<left>.+?)\s*\+\s*(?P<offset>-?\d+)\s*$")
@@ -391,3 +394,71 @@ def normalize_release_version(release: str) -> str: 
        parts = parts[:3]



    return ".".join(parts)





def normalize_working_group_alias(alias: str) -> WorkingGroup:

    """Normalize working group aliases to canonical working group enums.



    Supports: RAN, SA, CT and their common name variants.



    Args:

        alias: Working group alias or name



    Returns:

        Canonical working group enum value (WorkingGroup.RAN, WorkingGroup.SA, WorkingGroup.CT)



    Raises:

        ValueError: When the alias cannot be resolved

    """

    alias_upper = alias.strip().upper()

    cleaned = re.sub(r"[#\-_\s]", "", alias_upper)



    # Easy check via list(WorkingGroup): match against name or value

    for working_group in WorkingGroup:

        if cleaned.startswith(working_group.value) or cleaned.startswith(working_group.name):

            return working_group



    # If not found: match first letter

    if cleaned:

        for working_group in WorkingGroup:

            if working_group.value[0] == cleaned[0]:

                return working_group



    # If not found: any match of name/value in cleaned

    for working_group in WorkingGroup:

        if working_group.value in cleaned or working_group.name in cleaned:

            return working_group



    raise ValueError(f"Unknown working group: {alias}")





def normalize_subgroup_alias(alias: str) -> SubWorkingGroup:

    """Normalize subgroup aliases to canonical subgroup enums.



    Returns a SubWorkingGroup enum value (e.g., SubWorkingGroup.S4).

    Supports:

    - Short codes: S4, R1, C3

    - Full names: SA4, RAN1, CT3

    - Separators: SA-4, SA#4, SA 4

    - Plenary: RP, SP, CP, RAN PLENARY, etc.



    Args:

        alias: Subgroup alias or code



    Returns:

        Canonical subgroup enum value (e.g., SubWorkingGroup.S4)

    """

    input_str = alias.strip().upper()

    cleaned = re.sub(r"[#\-_\s]", "", input_str)



    # determine WorkingGroup first

    wg = normalize_working_group_alias(cleaned)



    # check if cleaned contains a _single_ number or the letter "P"

    match = re.search(r"(P|\d+)", cleaned)

    if not match:

        raise ValueError(f"Cannot parse/normalize subgroup format: {alias}")

    elif len(match.groups()) != 1:

        raise ValueError(f"Multiple/ambiguous matches found when parsing subgroup: {alias}")



    return SubWorkingGroup.from_wg_and_nbr(wg, match.group(1))  # e.g., S4, R1, C3