eth2.0-specs/scripts/function_puller.py

77 lines
3.4 KiB
Python
Raw Normal View History

2019-05-18 08:42:04 +00:00
import re
from typing import Dict, Tuple, NewType
2019-05-18 08:42:04 +00:00
FUNCTION_REGEX = r'^def [\w_]*'
2019-05-24 14:51:21 +00:00
BEGIN_INSERT_REGEX = r'# begin insert '
END_INSERT_REGEX = r'# end insert'
2019-05-18 08:42:04 +00:00
SpecObject = NewType('SpecObjects', Tuple[Dict[str, str], Dict[str, str], Dict[str, str], Dict[str, str]])
2019-05-18 08:42:04 +00:00
def get_spec(file_name: str) -> SpecObject:
"""
Takes in the file name of a spec.md file, opens it and returns the following objects:
functions = {function_name: function_code}
constants= {constant_name: constant_code}
ssz_objects= {object_name: object}
inserts= {insert_tag: code to be inserted}
Note: This function makes heavy use of the inherent ordering of dicts,
if this is not supported by your python version, it will not work.
"""
2019-05-24 14:51:21 +00:00
pulling_from = None # line number of start of latest object
current_name = None # most recent section title
insert_name = None # stores the label of the current insert object
2019-05-20 12:00:54 +00:00
functions = {}
2019-05-18 08:42:04 +00:00
constants = {}
2019-05-20 12:00:54 +00:00
ssz_objects = {}
2019-05-24 14:51:21 +00:00
inserts = {}
2019-05-18 08:42:04 +00:00
function_matcher = re.compile(FUNCTION_REGEX)
2019-05-24 14:51:21 +00:00
inserts_matcher = re.compile(BEGIN_INSERT_REGEX)
for linenum, line in enumerate(open(file_name).readlines()):
line = line.rstrip()
if pulling_from is None and len(line) > 0 and line[0] == '#' and line[-1] == '`':
2019-03-19 03:39:19 +00:00
current_name = line[line[:-1].rfind('`') + 1: -1]
if line[:9] == '```python':
assert pulling_from is None
pulling_from = linenum + 1
elif line[:3] == '```':
2019-05-18 08:42:04 +00:00
pulling_from = None
2019-05-24 14:51:21 +00:00
elif inserts_matcher.match(line) is not None:
# Find @insert names
2019-05-24 14:51:21 +00:00
insert_name = re.search(r'@[\w]*', line).group(0)
elif insert_name is not None:
# In insert mode, either the next line is more code, or the end of the insert
2019-05-24 14:51:21 +00:00
if re.match(END_INSERT_REGEX, line) is not None:
insert_name = None
else:
inserts[insert_name] = inserts.get(insert_name, '') + line + '\n'
else:
# Handle function definitions & ssz_objects
2019-05-18 08:42:04 +00:00
if pulling_from is not None:
2019-05-24 14:51:21 +00:00
func_match = function_matcher.match(line)
if func_match is not None:
current_name = func_match.group(0)
if function_matcher.match(current_name) is None: # The current line is an SSZ Object
2019-05-20 12:00:54 +00:00
ssz_objects[current_name] = ssz_objects.get(current_name, '') + line + '\n'
else: # The current line is code
2019-05-20 12:00:54 +00:00
functions[current_name] = functions.get(current_name, '') + line + '\n'
2019-05-18 08:42:04 +00:00
# Handle constant table entries
elif pulling_from is None and len(line) > 0 and line[0] == '|':
row = line[1:].split('|')
if len(row) >= 2:
for i in range(2):
row[i] = row[i].strip().strip('`')
if '`' in row[i]:
row[i] = row[i][:row[i].find('`')]
eligible = True
if row[0][0] not in 'ABCDEFGHIJKLMNOPQRSTUVWXYZ_':
eligible = False
for c in row[0]:
if c not in 'ABCDEFGHIJKLMNOPQRSTUVWXYZ_0123456789':
eligible = False
if eligible:
2019-05-18 08:42:04 +00:00
constants[row[0]] = row[1].replace('**TBD**', '0x1234567890123456789012345678901234567890')
2019-05-24 14:51:21 +00:00
return functions, constants, ssz_objects, inserts