mirror of
git://nv-tegra.nvidia.com/linux-nvgpu.git
synced 2025-12-22 09:12:24 +03:00
gpu: nvgpu: rfr: Add address book
Add an address book that lets devs use short hand for specifying --to and --cc targets. For example, if a dev wants to CC the MISRA list this can be used: $ ./scripts/rfr -e --cc misra ... The address book also lets devs add their own names/email addresses since this makes it convenient to CC individual people. For example: $ ./scripts/rfr -e --cc alex ... Several new arguments were added to support the address book. There are arguments to list/search the address book, ignore the address book, and to prevent nvgpu-core from being added to the to address by default. For more details see the help page. To use an address book there's several options: place one at ~/.rfr-addrbook Export an RFR_ADDRBOOK environment variable pointing to the address book, or specify one with the `-a' option. The address book contents is simple. All empty lines and lines beginning with '#' are ignored. The remaining lines are split by '|' and the first half of the line is considered a nickname and the latter half the address. An example: alex | alex waterman <alexw@nvidia.com> This will let you specify `--to alex' instead of the full email address. This is especially useful for mailing lists. Lastly there is more documentation located at: https://confluence.nvidia.com/display/TGS/NVGPU+Request+For+Review [Bump version to 1.1.0] Change-Id: Iac7ec05ae28d7e888d2bf36bd23574ec49eb04dc Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1983695 Reviewed-by: Rohit Khanna <rokhanna@nvidia.com>
This commit is contained in:
committed by
Rohit Khanna
parent
164e387940
commit
dd4c60aeb5
191
scripts/rfr_addrbook.py
Normal file
191
scripts/rfr_addrbook.py
Normal file
@@ -0,0 +1,191 @@
|
||||
# Copyright (c) 2019, NVIDIA CORPORATION. All Rights Reserved.
|
||||
#
|
||||
# Nvgpu address book. Entries in here let us do translations from human readable
|
||||
# email names/aliases to the real NV email addresses.
|
||||
#
|
||||
# This is structured as a dictionary with keys as human readable names which
|
||||
# point to the real email address.
|
||||
#
|
||||
# It's fine to have multiple keys pointing to the same thing. Aliases are nice.
|
||||
#
|
||||
|
||||
import re
|
||||
import os
|
||||
|
||||
# The table itself is private. Use the methods below to do AB lookups, etc.
|
||||
#
|
||||
# Also please note: the table should be all lower case! This makes it easy to
|
||||
# just lowercase all incoming queries so that we can do case insensitive
|
||||
# lookups.
|
||||
__rfr_address_book = { }
|
||||
|
||||
def __rfr_parse_addrbook(ab):
|
||||
"""
|
||||
Parse an address book. Ignore empty lines and lines that begin with '#'.
|
||||
All other lines are split by the '|' character into 2 strings - a key and
|
||||
a value. The key is a nickname and the value is the real address. For
|
||||
example:
|
||||
|
||||
alex | Alex Waterman <alexw@nvidia.com>
|
||||
|
||||
Note: if there's a syntax error detected an error is always printed. Unlike
|
||||
if there's missing files in which errors may be silent.
|
||||
|
||||
Returns True/False for pass/fail.
|
||||
"""
|
||||
|
||||
global __rfr_address_book
|
||||
|
||||
tmp_ab = { }
|
||||
line_nr = 0
|
||||
|
||||
for line in ab.readlines():
|
||||
line_nr += 1
|
||||
l = line.strip()
|
||||
|
||||
if len(l) == 0 or l[0] == '#':
|
||||
continue
|
||||
|
||||
# Only allow at most 1 split.
|
||||
kv = l.split('|', 1)
|
||||
|
||||
if len(kv) != 2:
|
||||
print('Error: unable to parse address book. Invalid line:')
|
||||
print(' > \'%s\' @ line: %d' % (line, line_nr))
|
||||
return False
|
||||
|
||||
tmp_ab[kv[0].strip().lower()] = kv[1].strip().lower()
|
||||
|
||||
__rfr_address_book = tmp_ab
|
||||
return True
|
||||
|
||||
def __rfr_load_ab(path, silent=False):
|
||||
"""
|
||||
Load the passed address book and print errors if silent is False.
|
||||
"""
|
||||
|
||||
success = True
|
||||
|
||||
try:
|
||||
with open(path) as ab:
|
||||
if not __rfr_parse_addrbook(ab):
|
||||
success = False
|
||||
except Exception as err:
|
||||
success = False
|
||||
if not silent:
|
||||
print 'Error: %s' % err
|
||||
|
||||
# It's not a very helpful error message I suppose. Eh. We will get more
|
||||
# detail from the __rfr_parse_addrbook() call itself.
|
||||
if not success and not silent:
|
||||
print('Failed to parse AB: %s' % path)
|
||||
|
||||
return success
|
||||
|
||||
def rfr_ab_load(ab_path):
|
||||
"""
|
||||
Attempt to load an address book.
|
||||
|
||||
If ab_path is None then look for the book at the environment variable
|
||||
RFR_ADDRBOOK. If that's not found or fails to load then fall back to trying
|
||||
~/.rfr-addrbook. If that doesn't work just silently return True and we will
|
||||
have an empty address book.
|
||||
|
||||
If ab_path is not None then try to load the addr book from the passed path
|
||||
and if there's an error print an error message. This will return the result
|
||||
of the AB load.
|
||||
"""
|
||||
|
||||
if ab_path:
|
||||
return __rfr_load_ab(ab_path)
|
||||
|
||||
# Ok, if we are here, do the whole env logic thing.
|
||||
ab_path = os.getenv('RFR_ADDRBOOK')
|
||||
if ab_path:
|
||||
success = __rfr_load_ab(ab_path, silent=True)
|
||||
if success:
|
||||
return True
|
||||
|
||||
# Otherwise... Try ~/.rfr-addrbook
|
||||
ab_path = os.getenv('HOME') + '/.rfr-addrbook'
|
||||
if ab_path:
|
||||
success = __rfr_load_ab(ab_path, silent=True)
|
||||
if success:
|
||||
return True
|
||||
|
||||
# Well, it all failed. But we don't care.
|
||||
return True
|
||||
|
||||
def rfr_ab_query(regex):
|
||||
"""
|
||||
Print a list of keys that match the passed regex string.
|
||||
"""
|
||||
|
||||
matches = [ ]
|
||||
|
||||
p = re.compile(regex, re.IGNORECASE)
|
||||
|
||||
for k in __rfr_address_book.keys():
|
||||
m = p.search(k)
|
||||
if not m:
|
||||
continue
|
||||
|
||||
matches.append(k)
|
||||
|
||||
print('Address book query: %s' % regex)
|
||||
|
||||
if not matches:
|
||||
print('> No matches')
|
||||
return
|
||||
|
||||
for addr in sorted(matches):
|
||||
print('> %-15s | %s' % (addr, __rfr_address_book[addr]))
|
||||
|
||||
def rfr_ab_lookup_single(addr):
|
||||
"""
|
||||
Exact lookup into the address book but if the addr isn't found in the keys
|
||||
then also check the values. If the addr is not found in the values then
|
||||
return None.
|
||||
|
||||
This lets the rfr script either ignore the AB lookup failure or bail out.
|
||||
"""
|
||||
|
||||
# If there's no address book, just pass the addr through.
|
||||
if len(__rfr_address_book.keys()) == 0:
|
||||
return addr
|
||||
|
||||
lc_addr = addr.lower()
|
||||
|
||||
if lc_addr in __rfr_address_book:
|
||||
return __rfr_address_book[lc_addr]
|
||||
|
||||
if lc_addr in __rfr_address_book.values():
|
||||
# Return the orignal, un-lowercased.
|
||||
return addr
|
||||
|
||||
return None
|
||||
|
||||
def rfr_ab_lookup(addrs, ignore_missing):
|
||||
"""
|
||||
Take a list of addresses and look them up in the address book. Return a new
|
||||
list which contains the results of the lookups.
|
||||
|
||||
If ignore_missing is True then if there's an address book lookup failure
|
||||
just ignore it and pass the address through to the new list. If False then
|
||||
fail and bail (return None).
|
||||
"""
|
||||
|
||||
lookups = [ ]
|
||||
|
||||
for a in addrs:
|
||||
lookup = rfr_ab_lookup_single(a)
|
||||
if not lookup and not ignore_missing:
|
||||
print('Unknown address: %s' % a)
|
||||
return None
|
||||
|
||||
if lookup:
|
||||
lookups.append(lookup)
|
||||
else:
|
||||
lookups.append(a)
|
||||
|
||||
return lookups
|
||||
Reference in New Issue
Block a user