# (Be in -*- python -*- mode.)
#
# ====================================================================
# Copyright (c) 2006-2007 CollabNet.  All rights reserved.
#
# This software is licensed as described in the file COPYING, which
# you should have received as part of this distribution.  The terms
# are also available at http://subversion.tigris.org/license-1.html.
# If newer versions of this license are posted there, you may use a
# newer version instead, at your option.
#
# This software consists of voluntary contributions made by many
# individuals.  For exact contribution history, see the revision
# history and logs, available at http://cvs2svn.tigris.org/.
# ====================================================================

#                  #####################
#                  ## PLEASE READ ME! ##
#                  #####################
#
# This is a template for an options file that can be used to configure
# cvs2svn.  Many options do not have defaults, so it is easier to copy
# this file and modify what you need rather than creating a new
# options file from scratch.
#
# This file is in Python syntax, but you don't need to know Python to
# modify it.  But if you *do* know Python, then you will be happy to
# know that you can use arbitary Python constructs to do fancy
# configuration tricks.
#
# But please be aware of the following:
#
# * In many places, leading whitespace is significant in Python (it is
#   used instead of curly braces to group statements together).
#   Therefore, if you don't know what you are doing, it is best to
#   leave the whitespace as it is.
#
# * In normal strings, Python uses backslashes ("\") are used as an
#   escape character.  Therefore you need to be careful, especially
#   when specifying regular expressions or Windows filenames.  It is
#   recommended that you use "raw strings" for these cases.
#   Backslashes in raw strings are treated literally.  A raw string is
#   written by prefixing an "r" character to a string.  Example:
#
#       ctx.sort_executable = r'c:\windows\system32\sort.exe'
#
# Two identifiers will have been defined before this file is executed,
# and can be used freely within this file:
#
#     ctx -- a Ctx object (see cvs2svn_lib/context.py), which holds
#         many configuration options
#
#     run_options -- an instance of the OptionsFileRunOptions class
#         (see cvs2svn_lib/run_options.py), which holds some variables
#         governing how cvs2svn is run


# Import some modules that are used in setting the options:
import re

from cvs2svn_lib.boolean import *
from cvs2svn_lib import config
from cvs2svn_lib import changeset_database
from cvs2svn_lib.common import CVSTextDecoder
from cvs2svn_lib.log import Log
from cvs2svn_lib.project import Project
from cvs2svn_lib.svn_output_option import DumpfileOutputOption
from cvs2svn_lib.svn_output_option import ExistingRepositoryOutputOption
from cvs2svn_lib.svn_output_option import NewRepositoryOutputOption
from cvs2svn_lib.revision_manager import NullRevisionRecorder
from cvs2svn_lib.revision_manager import NullRevisionExcluder
from cvs2svn_lib.rcs_revision_manager import RCSRevisionReader
from cvs2svn_lib.cvs_revision_manager import CVSRevisionReader
from cvs2svn_lib.checkout_internal import InternalRevisionRecorder
from cvs2svn_lib.checkout_internal import InternalRevisionExcluder
from cvs2svn_lib.checkout_internal import InternalRevisionReader
from cvs2svn_lib.symbol_strategy import AllBranchRule
from cvs2svn_lib.symbol_strategy import AllTagRule
from cvs2svn_lib.symbol_strategy import BranchIfCommitsRule
from cvs2svn_lib.symbol_strategy import ExcludeRegexpStrategyRule
from cvs2svn_lib.symbol_strategy import ForceBranchRegexpStrategyRule
from cvs2svn_lib.symbol_strategy import ForceTagRegexpStrategyRule
from cvs2svn_lib.symbol_strategy import HeuristicStrategyRule
from cvs2svn_lib.symbol_strategy import UnambiguousUsageRule
from cvs2svn_lib.symbol_strategy import DefaultBasePathRule
from cvs2svn_lib.symbol_strategy import HeuristicPreferredParentRule
from cvs2svn_lib.symbol_strategy import SymbolHintsFileRule
from cvs2svn_lib.symbol_transform import ReplaceSubstringsSymbolTransform
from cvs2svn_lib.symbol_transform import RegexpSymbolTransform
from cvs2svn_lib.symbol_transform import NormalizePathsSymbolTransform
from cvs2svn_lib.property_setters import AutoPropsPropertySetter
from cvs2svn_lib.property_setters import CVSBinaryFileDefaultMimeTypeSetter
from cvs2svn_lib.property_setters import CVSBinaryFileEOLStyleSetter
from cvs2svn_lib.property_setters import CVSRevisionNumberSetter
from cvs2svn_lib.property_setters import DefaultEOLStyleSetter
from cvs2svn_lib.property_setters import EOLStyleFromMimeTypeSetter
from cvs2svn_lib.property_setters import ExecutablePropertySetter
from cvs2svn_lib.property_setters import KeywordsPropertySetter
from cvs2svn_lib.property_setters import MimeMapper
from cvs2svn_lib.property_setters import SVNBinaryFileKeywordsPropertySetter

# To choose the level of logging output, uncomment one of the
# following lines:
#Log().log_level = Log.WARN
#Log().log_level = Log.QUIET
Log().log_level = Log.NORMAL
#Log().log_level = Log.VERBOSE
#Log().log_level = Log.DEBUG


# There are several possible options for where to put the output of a
# cvs2svn conversion.  Please choose one of the following and adjust
# the parameters as necessary:

# Use this output option if you would like cvs2svn to create a new SVN
# repository and store the converted repository there.  The first
# argument is the path to which the repository should be written (this
# repository must not already exist).  The second (optional) argument
# allows a --fs-type option to be passed to "svnadmin create".  The
# third (optional) argument can be specified to set the
# --bdb-txn-nosync option on a bdb repository:
ctx.output_option = NewRepositoryOutputOption(
    r'./svnrepo', # Path to repository
    fs_type='fsfs', # Type of repository to create
    #bdb_txn_nosync=False, # For bsd repositories, this option can be added
    )

# Use this output option if you would like cvs2svn to store the
# converted CVS repository into an SVN repository that already exists.
# The argument is the filesystem path of an existing local SVN
# repository (this repository must already exist):
#ctx.output_option = ExistingRepositoryOutputOption(
#    r'/path/to/svnrepo', # Path to repository
#    )

# Use this type of output option if you want the output of the
# conversion to be written to a SVN dumpfile instead of committing
# them into an actual repository:
#ctx.output_option = DumpfileOutputOption(
#    dumpfile_path=r'/path/to/cvs2svn-dump', # Name of dumpfile to create
#    )


# Independent of the ctx.output_option selected, the following option
# can be set to True to suppress cvs2svn output altogether:
ctx.dry_run = False

# The following set of options specifies how the revision contents of
# the RCS files should be read.
#
# The default selection is InternalRevisionReader, which uses built-in
# code that reads the RCS deltas while parsing the files in
# CollectRevsPass.  This method is very fast but requires lots of
# temporary disk space.  The disk space is required for (1) storing
# all of the RCS deltas, and (2) during OutputPass, keeping a copy of
# the full text of every revision that still has a descendant that
# hasn't yet been committed.  Since this can includes multiple
# revisions of each file (i.e., on multiple branches), the required
# amount of temporary space can potentially be many times the size of
# a checked out copy of the whole repository.  Setting compress=True
# cuts the disk space requirements by about 50% at the price of
# increased CPU usage.  Using compression usually speeds up the
# conversion due to the reduced I/O pressure, unless --tmpdir is on a
# RAM disk.  This method does not expand CVS's "Log" keywords.
#
# The second possibility is RCSRevisionReader, which uses RCS's "co"
# program to extract the revision contents of the RCS files during
# OutputPass.  This option doesn't require any temporary space, but it
# is relatively slow because (1) "co" has to be executed very many
# times; and (2) "co" itself has to assemble many file deltas to
# compute the contents of a particular revision.  The constructor
# argument specifies how to invoke the "co" executable.
#
# The third possibility is CVSRevisionReader, which uses the "cvs"
# program to extract the revision contents out of the RCS files during
# OutputPass.  This option doesn't require any temporary space, but it
# is the slowest of all, because "cvs" is considerably slower than
# "co".  However, it works in some situations where RCSRevisionReader
# fails; see the HTML documentation of the "--use-cvs" option for
# details.  The constructor argument specifies how to invoke the "co"
# executable.
#
# Choose one of the following three groups of lines:
ctx.revision_recorder = InternalRevisionRecorder(compress=True)
ctx.revision_excluder = InternalRevisionExcluder()
ctx.revision_reader = InternalRevisionReader(compress=True)

#ctx.revision_recorder = NullRevisionRecorder()
#ctx.revision_excluder = NullRevisionExcluder()
#ctx.revision_reader = RCSRevisionReader(co_executable=r'co')

#ctx.revision_recorder = NullRevisionRecorder()
#ctx.revision_excluder = NullRevisionExcluder()
#ctx.revision_reader = CVSRevisionReader(cvs_executable=r'cvs')

# Set the name (and optionally the path) of some other executables
# required by cvs2svn:
ctx.svnadmin_executable = r'svnadmin'
ctx.sort_executable = r'sort'

# Change the following line to True if the conversion should only
# include the trunk of the repository (i.e., all branches and tags
# should be ignored):
ctx.trunk_only = True

# Change the following line to True if cvs2svn should delete a
# directory once the last file has been deleted from it:
ctx.prune = True

# How to convert author names, log messages, and filenames to unicode.
# The first argument to CVSTextDecoder is a list of encoders that are
# tried in order in 'strict' mode until one of them succeeds.  If none
# of those succeeds, then fallback_encoder is used in lossy 'replace'
# mode (if it is specified).  Setting a fallback encoder ensures that
# the encoder always succeeds, but it can cause information loss.
ctx.cvs_author_decoder = CVSTextDecoder(
    [
        #'latin1',
        #'utf8',
        'ascii',
        ],
    fallback_encoding='latin1'
    )
ctx.cvs_log_decoder = CVSTextDecoder(
    [
        #'latin1',
        #'utf8',
        #'ascii',
        ],
    fallback_encoding='latin1'
    )
# You might want to be especially strict when converting filenames to
# unicode (e.g., maybe not specify a fallback_encoding).
ctx.cvs_filename_decoder = CVSTextDecoder(
    [
        #'latin1',
        #'utf8',
        'ascii',
        ],
    #fallback_encoding='latin1'
    )

# Some CVS clients for MacOS store resource fork data into CVS along
# with the file contents itself by wrapping it all up in a container
# format called "AppleSingle".  Subversion currently does not support
# MacOS resource forks.  Nevertheless, sometimes the resource fork
# information is not necessary and can be discarded.  Set the
# following option to True if you would like cvs2svn to identify files
# whose contents are encoded in AppleSingle format, and discard all
# but the data fork for such files before committing them to
# Subversion.  (Please note that AppleSingle contents are identified
# by the AppleSingle magic number as the first four bytes of the file.
# This check is not failproof, so only set this option if you think
# you need it.)
ctx.decode_apple_single = False

# This option can be set to the name of a filename to which are stored
# statistics and conversion decisions about the CVS symbols.
ctx.symbol_info_filename = None
#ctx.symbol_info_filename = 'symbol-info.txt'

# cvs2svn uses "symbol strategy rules" to help decide how to handle
# CVS symbols.  The rules in a project's symbol_strategy_rules are
# applied in order, and each rule is allowed to modify the symbol.
# The result (after each of the rules has been applied) is used for
# the conversion.
#
# 1. A CVS symbol might be used as a tag in one file and as a branch
#    in another file.  cvs2svn has to decide whether to convert such a
#    symbol as a tag or as a branch.  cvs2svn uses a series of
#    heuristic rules to decide how to convert a symbol.  The user can
#    override the default rules for specific symbols or symbols
#    matching regular expressions.
#
# 2. cvs2svn is also capable of excluding symbols from the conversion
#    (provided no other symbols depend on them.
#
# 3. CVS does not record unambiguously the line of development from
#    which a symbol sprouted.  cvs2svn uses a heuristic to choose a
#    symbol's "preferred parents".
#
# The standard branch/tag/exclude StrategyRules do not change a symbol
# that has already been processed by an earlier rule, so in effect the
# first matching rule is the one that is used.

global_symbol_strategy_rules = [
    # It is possible to specify manually exactly how symbols should be
    # converted and what line of development should be used as the
    # preferred parent.  To do so, create a file containing the symbol
    # hints and enable the following option.
    #
    # The format of the hints file is described in the documentation
    # for the SymbolHintsFileRule class in
    # cvs2svn_lib/symbol_strategy.py.  The file output by the
    # --write-symbol-info (i.e., ctx.symbol_info_filename) option is
    # in the same format.  The simplest way to use this option is to
    # run the conversion through CollateSymbolsPass with
    # --write-symbol-info option, copy the symbol info and edit it to
    # create a hints file, then re-start the conversion at
    # CollateSymbolsPass with this option enabled.
    #SymbolHintsFileRule('symbol-hints.txt'),

    # To force all symbols matching a regular expression to be
    # converted as branches, add rules like the following:
    #ForceBranchRegexpStrategyRule(r'branch.*'),

    # To force all symbols matching a regular expression to be
    # converted as tags, add rules like the following:
    #ForceTagRegexpStrategyRule(r'tag.*'),

    # To force all symbols matching a regular expression to be
    # excluded from the conversion, add rules like the following:
    #ExcludeRegexpStrategyRule(r'unknown-.*'),

    # Usually you want this rule, to convert unambiguous symbols
    # (symbols that were only ever used as tags or only ever used as
    # branches in CVS) the same way they were used in CVS:
    UnambiguousUsageRule(),

    # If there was ever a commit on a symbol, then it cannot be
    # converted as a tag.  This rule causes all such symbols to be
    # converted as branches.  If you would like to resolve such
    # ambiguities manually, comment out the following line:
    BranchIfCommitsRule(),

    # Last in the list can be a catch-all rule that is used for
    # symbols that were not matched by any of the more specific rules
    # above.  (Assuming that BranchIfCommitsRule() was included above,
    # then the symbols that are still indeterminate at this point can
    # sensibly be converted as branches or tags.)  Include at most one
    # of these lines.  If none of these catch-all rules are included,
    # then the presence of any ambiguous symbols (that haven't been
    # disambiguated above) is an error:

    # Convert ambiguous symbols based on whether they were used more
    # often as branches or as tags:
    HeuristicStrategyRule(),
    # Convert all ambiguous symbols as branches:
    #AllBranchRule(),
    # Convert all ambiguous symbols as tags:
    #AllTagRule(),

    # This rule sets the SVN path of each symbol to the default value
    # for the project containing it, if a path hasn't been set
    # already.
    DefaultBasePathRule(),

    # The last rule is here to choose the preferred parent of branches
    # and tags, that is, the line of development from which the symbol
    # sprouts.
    HeuristicPreferredParentRule(),
    ]

# Specify a username to be used for commits generated by cvs2svn.  If
# this options is set to None then no username will be used for such
# commits:
ctx.username = None
#ctx.username = 'cvs2svn'

# ctx.svn_property_setters contains a list of rules used to set the
# svn properties on files in the converted archive.  For each file,
# the rules are tried one by one.  Any rule can add or suppress one or
# more svn properties.  Typically the rules will not overwrite
# properties set by a previous rule (though they are free to do so).
ctx.svn_property_setters.extend([
    # To read auto-props rules from a file, uncomment the following line
    # and specify a filename.  The boolean argument specifies whether
    # case should be ignored when matching filenames to the filename
    # patterns found in the auto-props file:
    #AutoPropsPropertySetter(
    #    r'/home/username/.subversion/config',
    #    ignore_case=True,
    #    ),

    # To read mime types from a file, uncomment the following line and
    # specify a filename:
    #MimeMapper(r'/etc/mime.types'),

    # Omit the svn:eol-style property from any files that are listed
    # as binary (i.e., mode '-kb') in CVS:
    CVSBinaryFileEOLStyleSetter(),

    # If the file is binary and its svn:mime-type property is not yet
    # set, set svn:mime-type to 'application/octet-stream'.
    CVSBinaryFileDefaultMimeTypeSetter(),

    # To try to determine the eol-style from the mime type, uncomment
    # the following line:
    #EOLStyleFromMimeTypeSetter(),

    # Choose one of the following lines to set the default
    # svn:eol-style if none of the above rules applied.  The argument
    # is the svn:eol-style that should be applied, or None if no
    # svn:eol-style should be set (i.e., the file should be treated as
    # binary).
    #
    # The default is to treat all files as binary unless one of the
    # previous rules has determined otherwise, because this is the
    # safest approach.  However, if you have been diligent about
    # marking binary files with -kb in CVS and/or you have used the
    # above rules to definitely mark binary files as binary, then you
    # might prefer to use 'native' as the default, as it is usually
    # the most convenient setting for text files.  Other possible
    # options: 'CRLF', 'CR', 'LF'.
    DefaultEOLStyleSetter(None),
    #DefaultEOLStyleSetter('native'),

    # Prevent svn:keywords from being set on files that have
    # svn:eol-style unset.
    SVNBinaryFileKeywordsPropertySetter(),

    # If svn:keywords has not been set yet, set it based on the file's
    # CVS mode:
    KeywordsPropertySetter(config.SVN_KEYWORDS_VALUE),

    # Set the svn:executable flag on any files that are marked in CVS as
    # being executable:
    ExecutablePropertySetter(),

    # Uncomment the following line to include the original CVS revision
    # numbers as file properties in the SVN archive:
    #CVSRevisionNumberSetter(),

    ])

# The directory to use for temporary files:
ctx.tmpdir = r'cvs2svn-tmp'

# To skip the cleanup of temporary files, uncomment the following
# option:
#ctx.skip_cleanup = True


# In CVS, it is perfectly possible to make a single commit that
# affects more than one project or more than one branch of a single
# project.  Subversion also allows such commits.  Therefore, by
# default, when cvs2svn sees what looks like a cross-project or
# cross-branch CVS commit, it converts it into a
# cross-project/cross-branch Subversion commit.
#
# However, other tools and SCMs have trouble representing
# cross-project or cross-branch commits.  (For example, Trac's Revtree
# plugin, http://www.trac-hacks.org/wiki/RevtreePlugin is confused by
# such commits.)  Therefore, we provide the following two options to
# allow cross-project/cross-branch commits to be suppressed.

# To prevent CVS commits from different projects from being merged
# into single SVN commits, change this option to False:
ctx.cross_project_commits = True

# To prevent CVS commits on different branches from being merged into
# single SVN commits, change this option to False:
ctx.cross_branch_commits = True


# By default, it is a fatal error for a CVS ",v" file to appear both
# inside and outside of an "Attic" subdirectory (this should never
# happen, but frequently occurs due to botched repository
# administration).  If you would like to retain both versions of such
# files, change the following option to True, and the attic version of
# the file will be left in an SVN subdirectory called "Attic":
ctx.retain_conflicting_attic_files = False

# Now use stanzas like the following to define CVS projects that
# should be converted.  The arguments are:
#
# - The filesystem path of the project within the CVS repository.
#
# - The path that should be used for the "trunk" directory of this
#   project within the SVN repository.  This is an SVN path, so it
#   should always use forward slashes ("/").
#
# - The path that should be used for the "branches" directory of this
#   project within the SVN repository.  This is an SVN path, so it
#   should always use forward slashes ("/").
#
# - The path that should be used for the "tags" directory of this
#   project within the SVN repository.  This is an SVN path, so it
#   should always use forward slashes ("/").
#
# - A list of symbol transformations that can be used to rename
#   symbols in this project.  Each entry is a tuple (pattern,
#   replacement), where pattern is a Python regular expression pattern
#   and replacement is the text that should replace the pattern.  Each
#   pattern is matched against each symbol name.  If the pattern
#   matches, then it is replaced with the corresponding replacement
#   text.  The replacement can include substitution patterns (e.g.,
#   r'\1' or r'\g<name>').  Typically you will want to use raw strings
#   (strings with a preceding 'r', like shown in the examples) for the
#   regexp and its replacement to avoid backslash substitution within
#   those strings.


# Create the default project (using ctx.trunk, ctx.branches, and ctx.tags):
run_options.add_project(
    r'cvsrep4/code',
    trunk_path='trunk',
    branches_path='branches',
    tags_path='tags',
    symbol_transforms=[
        #RegexpSymbolTransform(r'release-(\d+)_(\d+)',
        #                      r'release-\1.\2'),
        #RegexpSymbolTransform(r'release-(\d+)_(\d+)_(\d+)',
        #                      r'release-\1.\2.\3'),
        # Convert backslashes into forward slashes:
        ReplaceSubstringsSymbolTransform('\\','/'),
        # Eliminate leading, trailing, and repeated slashes:
        NormalizePathsSymbolTransform(),
        ],
    symbol_strategy_rules=[
        # Additional, project-specific symbol strategy rules can
        # be added here.
        ] + global_symbol_strategy_rules,
    )

# Add a second project, to be stored to projA/trunk, projA/branches,
# and projA/tags:
#run_options.add_project(
#    r'my/cvsrepo/projA',
#    trunk_path='projA/trunk',
#    branches_path='projA/branches',
#    tags_path='projA/tags',
#    symbol_transforms=[
#        #RegexpSymbolTransform(r'release-(\d+)_(\d+)',
#        #                      r'release-\1.\2'),
#        #RegexpSymbolTransform(r'release-(\d+)_(\d+)_(\d+)',
#        #                      r'release-\1.\2.\3'),
#        # Convert backslashes into forward slashes:
#        ReplaceSubstringsSymbolTransform('\\','/'),
#        # Eliminate leading, trailing, and repeated slashes:
#        NormalizePathsSymbolTransform(),
#        ],
#    symbol_strategy_rules=[
#        # Additional, project-specific symbol strategy rules can
#        # be added here.
#        ] + global_symbol_strategy_rules,
#    )

# Change this option to True to turn on profiling of cvs2svn (for
# debugging purposes):
run_options.profiling = False


# Should CVSItem -> Changeset database files be memory mapped?  In
# some tests, using memory mapping speeded up the overall conversion
# by about 5%.  But this option can cause the conversion to fail with
# an out of memory error if the conversion computer runs out of
# virtual address space (e.g., when running a very large conversion on
# a 32-bit operating system).  Therefore it is disabled by default.
# Uncomment the following line to allow these database files to be
# memory mapped.
#changeset_database.use_mmap_for_cvs_item_to_changeset_table = True