Skip to content

Commit

Permalink
add a generate option to fill in TODOs for everything missing
Browse files Browse the repository at this point in the history
  • Loading branch information
@justdave
justdave committed May 3, 2024
1 parent 9a2ad3d commit 434d93b
Show file tree
Hide file tree
Showing 2 changed files with 317 additions and 16 deletions.
168 changes: 152 additions & 16 deletions make_schema_doc.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -22,15 +22,19 @@
import time
import sys
import argparse
from pprint import pprint, pformat
from black import FileMode, format_str

import schema_remarks
import get_schema


class BzSchemaProcessingException(Exception):
def __init__(self, message):
def __init__(self, errors):
message = 'foobar' #'<br />\n'.join(errors)
super().__init__(message)
self.message = message
self.errors = errors

def __str__(self):
return '\n' + self.message
Expand Down Expand Up @@ -768,27 +772,29 @@ def get_versioned_tables(first, last):
global errors
errors = []
if not first in schema_remarks.version_order:
raise BzSchemaProcessingException("I don't know about version '%s'." % last)
raise BzSchemaProcessingException([f"I don't know about version '{last}'."])
if not last in schema_remarks.version_order:
raise BzSchemaProcessingException("I don't know about version '%s'." % last)
raise BzSchemaProcessingException([f"I don't know about version '{last}'."])
if not (
schema_remarks.version_order.index(last)
>= schema_remarks.version_order.index(first)
):
raise BzSchemaProcessingException(
"Version '%s' comes before version '%s'." % (last, first)
[f"Version '{last}' comes before version '{first}'."]
)
colours = {}
tr = {}
if not first in schema_remarks.version_schema_map:
raise BzSchemaProcessingException(
"I know version '%s' exists, but I seem to be missing the data for it."
% last
[
f"I know version '{first}' exists, but I seem to be missing the data for it."
]
)
if not last in schema_remarks.version_schema_map:
raise BzSchemaProcessingException(
"I know version '%s' exists, but I seem to be missing the data for it."
% last
[
f"I know version '{last}' exists, but I seem to be missing the data for it."
]
)
schema_name = schema_remarks.version_schema_map[first]
schema, errors = get_schema.get_schema(schema_name, errors)
Expand All @@ -811,8 +817,7 @@ def get_versioned_tables(first, last):
# if we have errors at this point, it's fatal, there's no point
# in letting make_versioned_schema spew a ton more of them.
if errors:
e = str.join('<br/>\n', errors)
raise BzSchemaProcessingException(e)
raise BzSchemaProcessingException(errors)
schema = make_versioned_schema(schemas, colours, tr)
stringify_schema(schema)
return (schema, tr, colours, tuple(bugzilla_versions), errors)
Expand Down Expand Up @@ -864,8 +869,7 @@ def make_tables(first, last):
header = process(schema_remarks.header, bv, dict)
footer = process(schema_remarks.footer, bv, dict)
if errors:
e = str.join('<br/>\n', errors)
raise BzSchemaProcessingException(e)
raise BzSchemaProcessingException(errors)
return (header, body, footer)

def make_body(first, last):
Expand Down Expand Up @@ -894,14 +898,126 @@ def test_schema_remarks(args):
try:
(header, body, footer) = make_tables(first, last)
except BzSchemaProcessingException as e:
message = e.message
message = message.replace('<br/>', '')
print(message)
print('\n'.join(e.errors))
sys.exit()
print("Succeeded!")


def validate_schema_remarks(*args, **kwargs):
class regex_in:
string: str
match: re.Match = None

def __init__(self, thestring):
self.string = thestring

def __eq__(self, other: str | re.Pattern):
if isinstance(other, str):
other = re.compile(other)
assert isinstance(other, re.Pattern)
# TODO extend for search and match variants
self.match = other.fullmatch(self.string)
return self.match is not None

def __getitem__(self, group):
return self.match[group]


# Note the `as m` in in the case specification
# match regex_in(validated_string):
# case r'\d(\d)' as m:
# print(f'The second digit is {m[1]}')
# print(f'The whole match is {m.match}')


def generate_schema_remarks(args):
first = args.first
last = args.last
if last == None:
last = first
print(f"generating missing remarks for {first} .. {last}")
try:
(header, body, footer) = make_tables(first, last)
except BzSchemaProcessingException as e:
for error in e.errors:
match regex_in(error):
case r"No column remarks for table '(\S+)'\." as m:
if not m[1] in schema_remarks.table_remark:
schema_remarks.table_remark[m[1]] = 'TODO'
schema_remarks.column_remark[m[1]] = {}
case r"Table '(\S+)' has no remark for column '(\S+)'\." as m:
schema_remarks.column_remark[m[1]][m[2]] = 'TODO'
case r"No index remarks for table '(\S+)'\." as m:
schema_remarks.index_remark[m[1]] = {}
case r"Table '(\S+)' has no remark for index '(\S+)'\." as m:
schema_remarks.index_remark[m[1]][m[2]] = 'TODO'
case _:
print(f"Unhandled error: {error}")
# pprint(schema_remarks.column_remark)
if first != last:
# we're comparing two versions, run it a second time to catch added/removed
try:
(header, body, footer) = make_tables(first, last)
except BzSchemaProcessingException as e:
for error in e.errors:
match regex_in(error):
case r"No remark to add table (\S+)" as m:
schema_remarks.table_added_remark[m[1]] = 'TODO'
case r"No remark to remove table (\S+)" as m:
schema_remarks.table_removed_remark[m[1]] = 'TODO'
case r"No remark to add column (\S+)\.(\S+)\." as m:
if not m[1] in schema_remarks.column_added_remark:
schema_remarks.column_added_remark[m[1]] = {}
schema_remarks.column_added_remark[m[1]][m[2]] = 'TODO'
case r"No remark to remove column (\S+)\.(\S+)\." as m:
if not m[1] in schema_remarks.column_removed_remark:
schema_remarks.column_removed_remark[m[1]] = {}
schema_remarks.column_removed_remark[m[1]][m[2]] = 'TODO'
case r"No remark to add index (\S+):(\S+)\." as m:
if not m[1] in schema_remarks.index_added_remark:
schema_remarks.index_added_remark[m[1]] = {}
schema_remarks.index_added_remark[m[1]][m[2]] = 'TODO'
case r"No remark to remove index (\S+):(\S+)\." as m:
if not m[1] in schema_remarks.index_removed_remark:
schema_remarks.index_removed_remark[m[1]] = {}
schema_remarks.index_removed_remark[m[1]][m[2]] = 'TODO'
case _:
print(f"Unhandled error: {error}")
var_dict = {
'version_order': format_str('version_order = %s' % pformat(schema_remarks.version_order), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'default_first_version': format_str('default_first_version = %s' % pformat(schema_remarks.default_first_version), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'default_last_version': format_str('default_last_version = %s' % pformat(schema_remarks.default_last_version), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'version_schema_map': format_str('version_schema_map = %s' % pformat(schema_remarks.version_schema_map), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'version_remark': format_str('version_remark = %s' % pformat(schema_remarks.version_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'table_remark': format_str('table_remark = %s' % pformat(schema_remarks.table_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'table_added_remark': format_str('table_added_remark = %s' % pformat(schema_remarks.table_added_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'table_removed_remark': format_str('table_removed_remark = %s' % pformat(schema_remarks.table_removed_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'column_remark': format_str('column_remark = %s' % pformat(schema_remarks.column_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'column_renamed': format_str('column_renamed = %s' % pformat(schema_remarks.column_renamed), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'column_added_remark': format_str('column_added_remark = %s' % pformat(schema_remarks.column_added_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'column_removed_remark': format_str('column_removed_remark = %s' % pformat(schema_remarks.column_removed_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'index_remark': format_str('index_remark = %s' % pformat(schema_remarks.index_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'index_renamed': format_str('index_renamed = %s' % pformat(schema_remarks.index_renamed), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'index_removed_remark': format_str('index_removed_remark = %s' % pformat(schema_remarks.index_removed_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'index_added_remark': format_str('index_added_remark = %s' % pformat(schema_remarks.index_added_remark), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'notation_guide': format_str('notation_guide = %s' % pformat(schema_remarks.notation_guide), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'header': format_str('header = %s' % pformat(schema_remarks.header), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'footer': format_str('footer = %s' % pformat(schema_remarks.footer), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'prelude': format_str('prelude = %s' % pformat(schema_remarks.prelude), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
'afterword': format_str('afterword = %s' % pformat(schema_remarks.afterword), mode=FileMode(string_normalization=False,experimental_string_processing=True)),
}

with open('schema_remarks_template.txt', 'r') as infile:
template = infile.read()
with open('schema_remarks_new.py', 'w') as outfile:
output = template.format(**var_dict)
outfile.write(output)
print("Wrote changes to schema_remarks_new.py.")
print("diff the changes from schema_remarks.py and if you like them, move it overtop of it.")
sys.exit()
print("No changes detected.")


def validate_schema_remarks(args):
for v in schema_remarks.version_order:
if not v in schema_remarks.version_schema_map:
errors.append(
Expand Down Expand Up @@ -979,6 +1095,26 @@ def validate_schema_remarks(*args, **kwargs):
help="A file to write the generated schema doc to. Passing - will write it to standard out.",
)
parser_test.set_defaults(func=test_schema_remarks)
parser_generate = subparsers.add_parser(
'generate',
help='Add all of the missing remarks from a new schema to schema_remarks.py for you as TODO items.',
description='Add all of the missing remarks from a new schema to schema_remarks.py for you to as TODO items. Saves you the trouble of searching through the massive file looking for the right spot in alphabetical order to put them. Two benefits: gets a schema live faster (just without things documented), and you can search for TODO in the file to find the things that need updating.',
)
parser_generate.add_argument(
'first',
metavar="first",
choices=schema_remarks.version_order,
help="The starting version of the schemas to compare, or the single version to display if 'last' is not provided.",
)
parser_generate.add_argument(
'last',
metavar="last",
choices=schema_remarks.version_order,
nargs="?",
default=None,
help="The destination version of the schema to compare",
)
parser_generate.set_defaults(func=generate_schema_remarks)
args = parser.parse_args()
args.func(args)

Expand Down
Loading

0 comments on commit 434d93b

Please sign in to comment.