modseccfg: Check-in [f0887760c8]

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview

Comment:	Bundle logfmt1 into sub project. Support for /usr/share/logfmt/ database and update scripts (apache version is a trimmed down modseccfg.vhosts extractor). Support both pip and deb package (differ to some extend).
Downloads:	Tarball \| ZIP archive \| SQL archive
Timelines:	family \| ancestors \| descendants \| both \| trunk
Files:	files \| file ages \| folders
SHA3-256:	f0887760c86cf6ee08a7961533fbca4e7877ff11a69ef2c2f366a584d8196512
User & Date:	mario 2020-12-16 10:35:59

Context

2020-12-16
10:39		Use prefix/whitespace prepending for whole block (some macros just got the first line indented). check-in: a58faea2e0 user: mario tags: trunk
10:35		Bundle logfmt1 into sub project. Support for /usr/share/logfmt/ database and update scripts (apache version is a trimmed down modseccfg.vhosts extractor). Support both pip and deb package (differ to some extend). check-in: f0887760c8 user: mario tags: trunk
2020-12-15
09:41		Support TransferLog (which nobody uses). Add pretty ▰▰▰▰▰▰▰▰▱▱▱▱▱ progressbar for *.conf traversal on startup. check-in: d7f672369e user: mario tags: trunk

Changes

Hide Diffs Unified Diffs Ignore Whitespace Patch

Added logfmt1/README.md.

Added logfmt1/__init__.py.

Added logfmt1/grok2fmt1.

Name change from modseccfg/scripts/logex.py to logfmt1/logex.py.

Name change from modseccfg/logfmt1.py to logfmt1/logfmt1.py.

Added logfmt1/setup.py.

Added logfmt1/share/apache.combined.fmt.

Added logfmt1/share/apache.fmt.

Added logfmt1/share/grok.fmt.

Added logfmt1/share/strftime.fmt.

Added logfmt1/share/update/apache.

Added logfmt1/update-logfmt.

Added logfmt1/update_logfmt_all.py.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 ~~20 21 22 23 24~~ 25 26 27 28 29 30 31 ~~32 33 34 35~~ 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 ~~58 59~~ 60 61 62 63 64 65 66	#!/usr/bin/env python3 # encoding: utf-8 # title: logex # description: extract fields from log (with .fmt) # version: 0.2 # type: cli # category: extract # # # First parameter should be the log file. And a .log.fmt must exist ~~# alongside (see m~~odseccfg→Install~~~~ # # Syntax: ~~# logex.py /var/log/apache2/access.log @request_path @request_time @host~~ # # Other args: # --json / --tab / --csv # --iso8601 / --debug # # Where ~~the~~ @'s are decoration, and fields can be supplied as individual # arguments (~~then the output beca~~me space-separated). Field names are # application-type specific (internal) names. (e.g. @request_method, @host # or @tm_wday for Apache logs. With some predefined aliases, e.g. ~~for~~ w3c # field names.) # ~~# ~~Other f~~ield name prefixes ~~(than @)~~ are % or * to comma~~-join list~~ entries,~~ ~~# whi~~le +~~ combine~~s them~~ with ~~a plus, and #~~ made t~~he list field json.~~~~ # import sys, re, json import traceback, dateutil.parser ~~try:~~ import logfmt1 ~~except:~~ ~~from modseccfg import logfmt1~~ #-- args argv = sys.argv space = " " if "--tab" in argv: space = "\t" if "--csv" in argv: space = "," iso8601 = any(a in argv for a in ("--iso", "--iso8601", "--date", "--fixdates", "--die-apachedateformat-die")) as_json = any(a in argv for a in ("--json", "--asjson", "--as-json")) ~~dodebug = any(a in argv for a in ("--debug", "-D")) ~~or True~~~~ # remove --params ~~args = [a for a in argv if not re.match("^--\w+", a)]~~ # filename and field list log_fn = argv[1] output_fields = space.join(argv[2:]) #-- open log file try: reader = logfmt1.logopen(log_fn, debug=dodebug, duplicate=False) ~~if dodebug: sys.stdout.write(json.dumps(reader.__dict__, indent=2, default=lambda x:str(x))+"\n")~~ except Exception as e: sys.stderr.write(traceback.format_exc()+"\n") sys.stderr.write("Use `update-logfmt-apache` or modseccfg→File→Install→update_logfmt to generate a .fmt descriptor\n") sys.exit() # extra aliases (for apache/httpd)	\| \| \| \| \| \| \| \| > > > > > > > > > > \| > > > < \| < < \| \| \| \|	1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76	#!/usr/bin/env python3 # encoding: utf-8 # title: logex # description: extract fields from log (with .fmt) # version: 0.2 # type: cli # category: extract # # # First parameter should be the log file. And a .log.fmt must exist # alongside (generate with `update-logfmt`). # # Syntax: # logex.py /var/log/apache2/access.log request_path request_time @host # # Other args: # --json / --tab / --csv # --iso8601 / --debug # # Where any @'s are decoration, and fields can be supplied as individual # arguments (become space-separated without --tab/--csv). Field names are # application-type specific (internal) names. (E.g. @request_method, @host # or @tm_wday for Apache logs. With some predefined aliases, e.g. the w3c # extended log field names.) # # Field name prefixes are irrelevant for normal log entries. # But may join list-entries from container fields: # @name will just show the first entry # %name space-separated list # name comma-separated list # +name plus-joined list # #name as json array # name whatever # # Fields can be given as individual arguments, or as part of a string # output groups: # logex fn.log --tab @individual "@combined,@with,@comma" @tabagain # logex fn.log --csv "@lone" "multi" "#json" # Though you usually don't wanna overcomplicate the log format again. # import sys, re, json import traceback, dateutil.parser import logfmt1 #-- args argv = sys.argv space = " " if "--tab" in argv: space = "\t" if "--csv" in argv: space = "," iso8601 = any(a in argv for a in ("--iso", "--iso8601", "--date", "--fixdates", "--die-apachedateformat-die")) as_json = any(a in argv for a in ("--json", "--asjson", "--as-json")) dodebug = any(a in argv for a in ("--debug", "-D")) # remove --params argv = [a for a in argv if not re.match("^--\w+$\|^-\w$", a)] # filename and field list log_fn = argv[1] output_fields = space.join(argv[2:]) #-- open log file try: reader = logfmt1.logopen(log_fn, debug=dodebug, duplicate=False) #if dodebug: # sys.stdout.write(json.dumps(reader.__dict__, indent=2, default=lambda x:str(x))+"\n") except Exception as e: sys.stderr.write(traceback.format_exc()+"\n") sys.stderr.write("Use `update-logfmt-apache` or modseccfg→File→Install→update_logfmt to generate a .fmt descriptor\n") sys.exit() # extra aliases (for apache/httpd)
︙			︙
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98	} alias.update(reader.alias) # substitute occurences def get_field(m, row): pfx, name = m.groups() ~~val = row.get(alias.get(name~~, name),~~ "-")~~ if isinstance(val, list): # how to handle lists (for unpacked [key "value"] fields) ~~if pfx == "+":~~ val = "+".join(val) ~~elif pfx == "*" ~~or pfx == "%"~~:~~ val = ",".join(val) elif pfx == "#": val = json.dumps(val) else: val = str(val) return val	\| > > \| > > \|	90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112	} alias.update(reader.alias) # substitute occurences def get_field(m, row): pfx, name = m.groups() val = row.get(name) or row.get(alias.get(name)) or "-" if isinstance(val, list): # how to handle lists (for unpacked [key "value"] fields) if pfx == "@": val = val[0] elif pfx == "+": val = "+".join(val) elif pfx == "%": val = " ".join(val) elif pfx == "*": val = ",".join(val) elif pfx == "#": val = json.dumps(val) else: val = str(val) return val
︙			︙

1 2 ~~3 4~~ 5 6 7 8 9 10 11 12 13 14 15	# encoding: utf-8 # api: python ~~# title: logfmt1 # description: ~~craft~~ *.log.fmt ~~regex~~~~ # type: transform # category: io # version: 0.4 # license: Apache-2.0 # # Logging format strings to regex conversion. # # This is supposed to recognize .fmt files adjacent to logs, # which document both the application type and log variant # with the most current %p%l%ace%holder definition used. # The purpose of this is to allow log extraction with exact	\| \| > > > > > > > >	1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23	# encoding: utf-8 # api: python # title: python3-logfmt1 # description: handle *.log.fmt specifiers and regex conversion # type: transform # category: io # version: 0.4 # license: Apache-2.0 # pack: # logfmt1.py=/usr/lib/python3/dist-packages/ # update-logfmt=/usr/bin/ # ./logex.py=/usr/bin/logex # share=/usr/share/logfmt # architecture: all # depends: python (>= 3.6) # url: https://fossil.include-once.org/modseccfg/wiki/logfmt1 # # Logging format strings to regex conversion. # # This is supposed to recognize .fmt files adjacent to logs, # which document both the application type and log variant # with the most current %p%l%ace%holder definition used. # The purpose of this is to allow log extraction with exact
︙			︙
136 137 138 139 140 141 142 ~~143~~ 144 145 146 147 148 149 150 151 152 153 154 155 156 157 ~~158~~ 159 160 161 162 163 164 165	"alias": { "remote_address": "remote_addr", "ip": "remote_addr", "user": "remote_user", "file": "request_file", "size": "bytes_sent", "datetime": "request_time", ~~"ctime": "request_time".~~ "date": "request_time", "loglevel": "remote_logname", "module_name": "request_method", "request_flushed": "file_line", "requests_on_connection": "keepalives", "error": "apr_status", "request_flushed": "file_line", }, # convert variant placeholders into fields beforehand, # possibly looking up other definitions (strftime) for marshalled placeholders "expand": { "%\{([^{}]+)\}t": { "id": "request_time", ~~"class": "strftime" # different placeholders within \{...\}~~ }, "%[<>]?\{([\w\-]+)\}[Conexic]": { "id": "$1", "rx": "\S+" }, "%\{([\w\-]+)\}\^t[io]": { "id": "$1",	\| \| >	144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174	"alias": { "remote_address": "remote_addr", "ip": "remote_addr", "user": "remote_user", "file": "request_file", "size": "bytes_sent", "datetime": "request_time", "ctime": "request_time", "date": "request_time", "loglevel": "remote_logname", "module_name": "request_method", "request_flushed": "file_line", "requests_on_connection": "keepalives", "error": "apr_status", "request_flushed": "file_line", }, # convert variant placeholders into fields beforehand, # possibly looking up other definitions (strftime) for marshalled placeholders "expand": { "%\{([^{}]+)\}t": { "id": "request_time", "class": "strftime", # different placeholders within \{...\} "record": "$1" }, "%[<>]?\{([\w\-]+)\}[Conexic]": { "id": "$1", "rx": "\S+" }, "%\{([\w\-]+)\}\^t[io]": { "id": "$1",
︙			︙
232 233 234 235 236 237 238 ~~239 240 241 242~~ 243 244 245 246 247 248 249 250 251 ~~252~~ 253 254 255 256 257 258 259	"%%": { "id": "percent", "rx": "%" }, }, "expand": { "%(\w)": "[\w\d.]+" } } ~~~~grok = {~~ ~~# see dev/logfmt1/grok2fmt1.py~~ }~~ # return builtin definitions or from /usr/share/logfmt/..fmt @staticmethod def get(cls): rules = {} cls = cls.split(" ") while cls: lookup = ".".join(cls) lookup_ = "_".join(cls) add = {} ~~fn = f"/usr/share/log~~fmt/{lookup}.~~fmt"~~ if os.path.exists(fn): add = open(fn, "r", encoding="utf-8") add = re.sub("^\s#.+", "", add, re.M) add = json.loads(add) #elif .grok: read, find primary regex:, or declare %GROK per cls=["grok", "-"] #elif *.lnav: get readymade "regex:" else:	< < < < \| > > >	241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267	"%%": { "id": "percent", "rx": "%" }, }, "expand": { "%(\w)": "[\w\d.]+" } } # return builtin definitions or from /usr/share/logfmt/..fmt @staticmethod def get(cls): rules = {} cls = cls.split(" ") while cls: lookup = ".".join(cls) lookup_ = "_".join(cls) add = {} dir = "/usr/share/logfmt" if not os.path.exists(dir): # kludge for Python package dir = re.sub("[\w.]$", "share", __file__) # use bundled ./share/ directory fn = f"{dir}/{lookup}.fmt" if os.path.exists(fn): add = open(fn, "r", encoding="utf-8") add = re.sub("^\s#.+", "", add, re.M) add = json.loads(add) #elif .grok: read, find primary regex:, or declare %GROK per cls=["grok", "-"] #elif *.lnav: get readymade "regex:" else:
︙			︙
268 269 270 271 272 273 274 275 276 277 278 279 280 281	if isinstance(v, dict): if not k in rules: rules[k] = {} rulesdb.merge(rules[k], v) elif not k in rules: rules[k] = v return rules # should be the other way round: regex() is meant to be a subset of update() def update(fmt): fmt["regex"] = regex(fmt, update=True) # assemble regex for format string	> > > > > > >	276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296	if isinstance(v, dict): if not k in rules: rules[k] = {} rulesdb.merge(rules[k], v) elif not k in rules: rules[k] = v return rules #@staticmethod #def extract_all(): # for key,val in self.__dict__.items(): # if isinstance(val, dict): # open("share/{key}.fmt", "w").write(json.dumps(val, indent=4)) # should be the other way round: regex() is meant to be a subset of update() def update(fmt): fmt["regex"] = regex(fmt, update=True) # assemble regex for format string
︙			︙
307 308 309 310 311 312 313 ~~314 315~~ ~~316 317 318 319~~ ~~320~~ ~~321 322 323 324~~ ~~325 326 327 328 329 330~~ 331 332 333 334 335 336 337	if "rewrite" in rules: for rx, repl in rules["rewrite"].items(): record = re.sub(rx, repl.replace(r'$1', r'\1'), record) # create fields from variant placeholders if "expand" in rules: for rx, expand in rules["expand"].items(): ~~for is_quoted, ~~key, id~~, ~~extra~~ in re.findall(f"(\"?)({rx})", record): if ~~not key~~ in fields:~~ ~~~~opt~~ = copy(expand) ~~#subs = ["", key, id, extra] replace any $…~~ # $… ~~if opt~~["id"].find('$1') >= 0:~~ ~~~~opt~~["id"] = ~~opt["id"].replace('$1',~~ re.sub("\W+", "_", id).lower())~~ if not "rx" in ~~opt~~ and "class" in ~~opt~~: ~~# replace any $…~~ ~~opt~~["rx"] = regex({~~"class":opt["class"], "record":id})~~ ~~pass~~ if "rx" in ~~opt~~: if not is_quoted and ~~opt~~["rx"] == '[^"]': ~~opt~~["rx"] = "\S" elif is_quoted and ~~opt~~["rx"] == "\S+": ~~opt~~["rx"] = "(?:[^\"]*\|\\\\\")+" fields[~~key~~] = ~~opt~~ # catch-all \S+ for completely unknown placeholders if "placeholder" in rules: for ph in re.findall(rules["placeholder"], record): if not ph in fields: id = re.sub("\W+", "", ph) fields[ph] = { "id": id, "rx": "\S+" }	\| \| > \| < \| \| > \| > \| < \| \| > > > \| \| \| \| \| \|	322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356	if "rewrite" in rules: for rx, repl in rules["rewrite"].items(): record = re.sub(rx, repl.replace(r'$1', r'\1'), record) # create fields from variant placeholders if "expand" in rules: for rx, expand in rules["expand"].items(): for is_quoted, match, uu in re.findall(f"(\"?)({rx})", record): if match in fields: continue x = copy(expand) # id: is usually "$1", but might be "prefix_$2" or something if x["id"].find('$') >= 0: x["id"] = rx_sub(rx, x["id"], match) x["id"] = re.sub("\W+", "", x["id"]).lower() # recurse into other pattern types if not "rx" in x and "class" in x: x["rx"] = regex({ "class": x["class"], "record": rx_sub(rx, x.get("record") or "$1", match) }) # handle generic placeholders / quoted strings, somewhat apache-specific if "rx" in x: if not is_quoted and x["rx"] == '[^"]': x["rx"] = "\S" elif is_quoted and x["rx"] == "\S+": x["rx"] = "(?:[^\"]\|\\\\\")+" fields[match] = x # catch-all \S+ for completely unknown placeholders if "placeholder" in rules: for ph in re.findall(rules["placeholder"], record): if not ph in fields: id = re.sub("\W+", "", ph) fields[ph] = { "id": id, "rx": "\S+" }
︙			︙
349 350 351 352 353 354 355 ~~356~~ 357 358 359 360 361 362 363 364 365 366 367 368 ~~369~~ 370 371 372 ~~373~~ 374 375 376 377 378 379 380	else: rx = f"(?<{id}>{rx})" return rx rx = re.sub(rules["placeholder"], sub_placeholder, record) rx = rename_duplicates(rx) return rx # (?<duplicate2>) def rename_duplicates(rx): fields = [] def count_ph(m): s, i = m.group(1), 1 while s in fields: i += 1 s = f"{m.group(1)}{i}" fields.append(s) return s return re.sub("(?<=\(\?\<)(\w+)(?=\>)", count_ph, rx) # (?<name>) to (?P<name>) def rx2re(rx): return re.sub("\(\?<(?=\w+>)", "(?P<", rx) # file-style wrapper that yields parsed dictionaries instead of string lines class parsy_parse: def __init__(self, logfn="", fmt=None, debug=False, fail=False, duplicate=True): """	< < \| > > > > >	368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402	else: rx = f"(?<{id}>{rx})" return rx rx = re.sub(rules["placeholder"], sub_placeholder, record) rx = rename_duplicates(rx) return rx # (?<duplicate2>) def rename_duplicates(rx): fields = [] def count_ph(m): s, i = m.group(1), 1 while s in fields: i += 1 s = f"{m.group(1)}{i}" fields.append(s) return s return re.sub("(?<=\(\?\<)(\w+)(?=\>)", count_ph, rx) # (?<name>) to (?P<name>) def rx2re(rx): return re.sub("\(\?<(?=\w+>)", "(?P<", rx) # allow for $1, $2, $3 in re.sub() def rx_sub(pattern, replacement, source, flags=0): if replacement.find('$') >= 0: replacement = re.sub(r'\(?=[0-9])', '$', replacement) return re.sub(pattern, replacement, source, flags) # file-style wrapper that yields parsed dictionaries instead of string lines class parsy_parse: def __init__(self, logfn="", fmt=None, debug=False, fail=False, duplicate=True): """
︙			︙
427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 ~~451~~ 452 453 ~~454 455 456 457~~ ~~458 459 460~~ 461 462 463 464 465 466 ~~467~~ ~~468~~ 469 470 471 472 473 474 475 476	self.debug_rx(line) if self.fail: raise StopIteration() elif self.fail: raise StopIteration() else: pass # jusst try next line # add [key "value"] fields def container_expand(self, d): for k,opt in self.container.items(): if k in d: for id,val in re.findall(opt["rx"], d[k]): if not id in d: d[id] = val elif not isinstance(d[id], "list"): d[id] = [d[id], val] else: d[id].append(val) # ANSI output for debugging regex/fmt string def debug_rx(self, line): rx = self.rx.pattern line = line.rstrip() ~~rx_cut = re.compile("[\s\"\[]\(\?P<\w+>([^()]+\|\([^()]+\))\)~~[\"\]\s~~]*$")~~ # iteratively strip (?...) capture groups while len(rx) and rx.find("(?P<") >= 0: ~~fail = rx_cut.search(rx) ~~if fail:~~ fail = fail.group(0) ~~else:~~~~ ~~fail = "<unknown-last-capture>" ~~break~~ rx = rx~~_cu~~t~~.sub("", rx)~~~~ try: if re.match(rx, line): break # works now, so `fail` was the culprit except: # likely broke regex nesting, try removing next (?...) pass ~~matched = re.match(rx, line~~).group(0~~)~~ ~~print("\033[42m" +~~ matched ~~+ "\033[41m" + line[len(matched):] + "\033[40;0m")~~ print("\033[36m" + "failed regex section: \033[1;33;41m" + fail + "\033[40;0m") # alias logopen = parsy_parse #-- test def test():	> > > > \| \| < \| \| > > \| < \| > > > > \| > > \| >	449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509	self.debug_rx(line) if self.fail: raise StopIteration() elif self.fail: raise StopIteration() else: pass # jusst try next line # pass .close() and similar to file object def __getattr__(self, name): return getattr(self.f, name) # add [key "value"] fields def container_expand(self, d): for k,opt in self.container.items(): if k in d: for id,val in re.findall(opt["rx"], d[k]): if not id in d: d[id] = val elif not isinstance(d[id], "list"): d[id] = [d[id], val] else: d[id].append(val) # ANSI output for debugging regex/fmt string def debug_rx(self, line): rx = self.rx.pattern line = line.rstrip() #rx_cut = re.compile("[^)]* \(\?P<\w+> ( [^()]+ \| \([^()]+\) )+ \) [^()]* \Z", re.X) # iteratively strip (?...) capture groups while len(rx) and rx.find("(?P<") >= 0: #fail = rx_cut.search(rx) #if fail: fail = fail.group(0) #else: fail = "<unknown-last-capture>"; break last = rx.rindex("(?P<") if last < 1: fail = "<unknown-last-capture>"; break fail = rx[last:] #print(f"testfail: `{fail}`") try: rx = rx[0:last] rx = re.sub("[^)]*$", "", rx) if re.match(rx, line): break # works now, so `fail` was the culprit except: # likely broke regex nesting, try removing next (?...) pass try: matched = re.match(rx, line) matched = matched.group(0) except: matched = "" print("\033[36m" + "failed regex section: \033[1;33;41m" + fail + "\033[40;0m") print("\033[42m" + matched + "\033[41m" + line[len(matched):] + "\033[40;0m") # alias logopen = parsy_parse #-- test def test():
︙			︙