# AUTOGENERATED FILE! PLEASE DON'T EDIT
__all__ = ["grep", "grepTemplate"]
import re, k1lib
from k1lib.cli.init import BaseCli, Table, Row; import k1lib.cli as cli
from collections import deque; from typing import Iterator, Union, Callable, Any
inf = float("inf")
[docs]class grep(BaseCli):
[docs] def __init__(self, pattern:Union[str, Callable[[Any], bool]], before:int=0, after:int=0, N:int=float("inf"), sep:bool=False, col:int=None):
"""Find lines that has the specified pattern.
Example::
# returns ['d', 'd']
"abcde12d34" | grep("d") | deref()
# returns ['c', 'd', '2', 'd'], 2 sections of ['c', 'd'] and ['2', 'd']
"abcde12d34" | grep("d", 1) | deref()
# returns ['c', 'd']
"abcde12d34" | grep("d", 1, N=1) | deref()
# returns ['d', 'e', 'd', '3', '4'], 2 sections of ['d', 'e'] and ['d', '3', '4']
"abcde12d34" | grep("d", 0, 3).till("e") | deref()
# returns [['0', '1', '2'], ['3', '1', '4']]
"0123145" | grep("1", 2, 1, sep=True) | deref()
You can also separate out the sections::
# returns [['c', 'd'], ['2', 'd']]
"abcde12d34" | grep("d", 1, sep=True) | deref()
# returns [['c', 'd']]
"abcde12d34" | grep("d", 1, N=1, sep=True) | deref()
# returns [['1', '2', '3'], ['1', '4', '5']]
"0123145" | grep("1", sep=True).till() | deref()
You can also put in predicates instead of regex patterns::
# returns ['d', 'd']
"abcde12d34" | grep(lambda x: x == "d") | deref()
# also returns ['d', 'd']
"abcde12d34" | filt(lambda x: x == "d") | deref()
# returns ['d', 'e', 'd', '3', '4']
"abcde12d34" | grep(lambda x: x == "d").till(lambda x: x == "e") | deref()
The first scenario looks like a regular filter function, already implemented by :class:`~k1lib.cli.filt.filt`,
but :class:`grep` brings in more clustering features for the price of reduced
execution speed. So for simple scenarios it's advised that you use :class:`~k1lib.cli.filt.filt`.
See also: :class:`~k1lib.cli.structural.groupBy`
Also, there's a `whole tutorial <../tutorials/cli.html>`_ devoted to just this cli
:param pattern: regex pattern to search for in a line
:param before: lines before the hit. Outputs independent lines
:param after: lines after the hit. Outputs independent lines
:param N: max sections to output
:param sep: whether to separate out the sections as lists
:param col: searches for pattern in a specific column"""
super().__init__()
if isinstance(pattern, str):
self._f = re.compile(pattern).search # make func quickly accessible
else: self._f = cli.op.solidify(pattern)
self.before = before; self.after = after; self.col = col; self.N = N; self.sep = sep
self.tillPattern = None; self.tillAfter = None; self._tillF = lambda x: False
[docs] def till(self, pattern:Union[str, Callable[[Any], bool]]=None):
"""Greps until some other pattern appear. Inclusive, so you might want to
trim the last line. Example::
# returns ['5', '6', '7', '8'], includes last item
range(10) | join("") | grep("5").till("8") | deref()
# returns ['d', 'e', 'd', '3', '4']
"abcde12d34" | grep("d").till("e") | deref()
# returns ['d', 'e']
"abcde12d34" | grep("d", N=1).till("e") | deref()
If initial pattern and till pattern are the same, then you don't have use this method at
all. Instead, do something like this::
# returns ['1', '2', '3']
"0123145" | grep("1", after=1e9, N=1) | deref()"""
if pattern is None:
self._tillF = self._f
elif isinstance(pattern, str):
self._tillF = re.compile(pattern).search
else:
self._tillF = cli.op.solidify(pattern)
self.tillAfter = self.after; self.after = inf; return self
[docs] def __ror__(self, it:Iterator[str]) -> Iterator[str]:
self.sectionIdx = 0; col = self.col; _f = self._f; _tillF = self._tillF
if self.sep:
elems = []; idx = 0
s = self._clone(); s.sep = False
for line in (it | s):
if s.sectionIdx > idx: # outputs whatever remaining
if len(elems) > 0: yield list(elems)
idx = s.sectionIdx; elems = []
elems.append(line)
yield list(elems); return
queue = deque([], self.before); counter = 0 # remaining lines after to display
cRO = k1lib.RunOnce(); cRO.done()
for line in it:
if col != None: line = list(line); elem = line[col]
else: elem = line
if _f(elem): # new section
self.sectionIdx += 1; counter = self.after+1; cRO.revert()
if self.sectionIdx > self.N: return
yield from queue; queue.clear(); yield line
elif _tillF(elem) and counter == inf: # closing section
counter = self.tillAfter + 1; cRO.revert(); yield line
if counter == 0:
queue.append(line) # saves recent past lines
elif counter > 0: # yielding "after" section
if cRO.done(): yield line
counter -= 1
def _clone(self):
answer = grep(self._f, self.before, self.after, self.N, self.sep, self.col)
answer._tillF = self._tillF; answer.tillAfter = self.tillAfter; return answer
[docs]class grepTemplate(BaseCli):
[docs] def __init__(self, pattern:str, template:str):
"""Searches over all lines, pick out the match, and expands
it to the templateand yields"""
super().__init__()
self.pattern = re.compile(pattern); self.template = template
[docs] def __ror__(self, it:Iterator[str]):
super().__ror__(it)
for line in it:
matchObj = self.pattern.search(line)
if matchObj is None: continue
yield matchObj.expand(self.template)