TextFileProcessing.gmk

#
# Copyright (c) 2013, 2019, Oracle and/or its affiliates. All rights reserved.
# DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
#
# This code is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License version 2 only, as
# published by the Free Software Foundation.  Oracle designates this
# particular file as subject to the "Classpath" exception as provided
# by Oracle in the LICENSE file that accompanied this code.
#
# This code is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
# version 2 for more details (a copy is included in the LICENSE file that
# accompanied this code).
#
# You should have received a copy of the GNU General Public License version
# 2 along with this work; if not, write to the Free Software Foundation,
# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
# or visit www.oracle.com if you need additional information or have any
# questions.
#

ifeq (,$(_MAKEBASE_GMK))
  $(error You must include MakeBase.gmk prior to including TextFileProcessing.gmk)
endif

# Helper function for SetupTextFileProcessing; adds a rule for a single file
# to be processed.
# param 1 = The namespace argument, e.g. BUILD_VERSION_FILE
# param 2 = the source file name (full path)
# param 3 = the target base directory
# param 4 = the target file name (possibly with a partial path)
define SetupSingleTextFileForProcessing
  $(strip $3)/$(strip $4): $2 $$($1_VARDEPS_FILE)
	$$(call LogInfo, Processing $(strip $4))
	$$(call MakeDir, $$(@D))
	$(RM) '$$@' '$$@.includes.tmp' '$$@.replacements.tmp' ; \
	$$($1_INCLUDES_COMMAND_LINE) < '$$<' > '$$@.includes.tmp' ; \
	$$($1_REPLACEMENTS_COMMAND_LINE) < '$$@.includes.tmp' > '$$@.replacements.tmp' ; \
	$(RM) '$$@.includes.tmp' ; \
	$(MV) '$$@.replacements.tmp' '$$@'

  $1 += $(strip $3)/$(strip $4)
endef

# Setup make rules for processing one or more text files, in which specified
# markers are replaced with a given text, or with the contents of a given file.
#
# Parameter 1 is the name of the rule. This name is used as variable prefix,
# and the targets generated are listed in a variable by that name.
#
# Remaining parameters are named arguments. These include:
#   SOURCE_DIRS one or more directory roots to search for files to process
#   SOURCE_FILES complete paths to one or more files to process
#   OUTPUT_DIR the directory where we store the processed files.
#   OUTPUT_FILE the name of the resulting file. Only allowed if processing a
#       single file.
#   SOURCE_BASE_DIR a common root to all SOURCE_DIRS.
#       If specified, files will keep the path relative to the base in the
#       OUTPUT_DIR. Otherwise, the hierarchy will be flattened into the OUTPUT_DIR.
#   INCLUDE_FILES only include files matching these patterns (used only with
#       SOURCE_DIRS)
#   EXCLUDE_FILES exclude files matching these patterns (used only with
#       SOURCE_DIRS)
#   INCLUDES replace the occurances of a pattern with the contents of a file;
#       one or more such include pattern, using the syntax:
#       PLACEHOLDER => FILE_TO_INCLUDE ; ...
#       Each PLACEHOLDER must be on a single, otherwise empty line (whitespace
#       padding is allowed).
#   REPLACEMENTS one or more text replacement patterns, using the syntax:
#       PATTERN => REPLACEMENT_TEXT ; ...
#
#   If both INCLUDES or REPLACEMENTS are present, then the includes will be
#   processed first, and replacements will be done on the included fragments as well.
#   If neither is present, the files will just be copied without modifications.
#
SetupTextFileProcessing = $(NamedParamsMacroTemplate)
define SetupTextFileProcessingBody

  ifneq ($$($1_SOURCE_FILES),)
    ifneq ($$($1_SOURCE_DIRS),)
      $$(error Cannot use both SOURCE_FILES and SOURCE_DIRS (in $1))
    endif
    ifneq ($$($1_SOURCE_BASE_DIR),)
      $$(error Cannot use SOURCE_BASE_DIR without SOURCE_DIRS (in $1))
    endif
    ifneq ($$($1_EXCLUDE_FILES)$$($1_INCLUDE_FILES),)
      $$(error Cannot INCLUDE/EXCLUDE_FILES with SOURCE_FILES (in $1))
    endif
  else
    ifeq ($$($1_SOURCE_DIRS),)
      $$(error Must specify either SOURCE_FILES or SOURCE_DIRS (in $1))
    endif
    # Find all files in the source trees. Sort to remove duplicates.
    $$(foreach src, $$($1_SOURCE_DIRS), $$(if $$(wildcard $$(src)), , \
        $$(error SOURCE_DIRS contains missing directory $$(src) (in $1))))
    ifneq ($$($1_SOURCE_BASE_DIR),)
      $$(foreach src, $$($1_SOURCE_DIRS), \
          $$(if $$(findstring $$($1_SOURCE_BASE_DIR), $$(src)), , \
          $$(error SOURCE_DIRS contains directory $$(src) outside \
              SOURCE_BASE_DIR $$($1_SOURCE_BASE_DIR) (in $1))))
    endif
    $1_SOURCE_FILES := $$(sort $$(call FindFiles,$$($1_SOURCE_DIRS)))
    $1_EXCLUDE_FILES:=$$(foreach i,$$($1_SOURCE_DIRS),$$(addprefix $$i/,$$($1_EXCLUDE_FILES)))
    $1_INCLUDE_FILES:=$$(foreach i,$$($1_SOURCE_DIRS),$$(addprefix $$i/,$$($1_INCLUDE_FILES)))
    $1_SOURCE_FILES := $$(filter-out $$($1_EXCLUDE_FILES),$$($1_SOURCE_FILES))
    ifneq (,$$(strip $$($1_INCLUDE_FILES)))
      $1_SOURCE_FILES := $$(filter $$($1_INCLUDE_FILES),$$($1_SOURCE_FILES))
    endif
    ifeq (,$$($1_SOURCE_FILES))
      $$(info No sources found for $1 when looking inside the dirs $$($1_SRC))
    endif
  endif

  ifneq ($$($1_REPLACEMENTS),)
    # We have a replacement request, prepare it for the recipe
    ifneq ($$(findstring /,$$($1_REPLACEMENTS)),)
      # Cannot use / as separator
      ifneq ($$(findstring @,$$($1_REPLACEMENTS)),)
        # Cannot use @ as separator
        ifneq ($$(findstring |,$$($1_REPLACEMENTS)),)
          # Cannot use | as separator
          ifneq ($$(findstring !,$$($1_REPLACEMENTS)),)
            # Cannot use ! as separator. Give up.
            $$(error No suitable sed separator can be found for $1. Tested /, @, | and !)
          else
            $1_SEP := !
          endif
        else
          $1_SEP := |
        endif
      else
        $1_SEP := @
      endif
    else
      $1_SEP := /
    endif

    # If we have a trailing "=>" (i.e. last rule replaces with empty, and is not
    # terminated by a ;), add a trailing ; to minimize the number of corner
    # cases in the hairy subst expression..
    ifeq ($$(lastword $$($1_REPLACEMENTS)), =>)
      $1_REPLACEMENTS += ;
    endif

    # If we have a trailing ";", add a dummy replacement, since there is no easy
    # way to delete the last word in make.
    ifeq ($$(lastword $$($1_REPLACEMENTS)), ;)
      $1_REPLACEMENTS += DUMMY_REPLACEMENT => DUMMY_REPLACEMENT
    endif

    # Convert the REPLACEMENTS syntax ( A => B ; C => D ; ...) to a sed command
    # line (-e "s/A/B/g" -e "s/C/D/g" ...), basically by replacing '=>' with '/'
    # and ';' with '/g" -e "s/', and adjusting for edge cases.
    $1_REPLACEMENTS_COMMAND_LINE := $(SED) -e 's$$($1_SEP)$$(subst $$(SPACE);$$(SPACE),$$($1_SEP)g' \
        -e 's$$($1_SEP),$$(subst $$(SPACE)=>$$(SPACE),$$($1_SEP),$$(subst $$(SPACE)=>$$(SPACE);$$(SPACE),$$($1_SEP)$$($1_SEP)g' \
        -e 's$$($1_SEP),$$(strip $$($1_REPLACEMENTS)))))$$($1_SEP)g'
  else
    # We don't have any replacements, just pipe the file through cat.
    $1_REPLACEMENTS_COMMAND_LINE := $(CAT)
  endif

  ifneq ($$($1_INCLUDES),)
    # We have a include request, prepare it for the recipe.
    # Convert an INCLUDE like this PATTERN_1 => file1 ; PATTERN_2 => file2 ;
    # into an awk script fragment like this:
    # {
    #   if (matches("PATTERN_1")) { include("file1") } else
    #   if (matches("PATTERN_2")) { include("file2") } else
    #   print
    # }

    $1_INCLUDES_HEADER_AWK := \
        function matches(pattern) { return ($$$$0 ~ "^[ \t]*" pattern "[ \t]*$$$$") } \
        function include(filename) { while ((getline < filename) == 1) print ; close(filename) }
    $1_INCLUDES_PARTIAL_AWK := $$(subst $$(SPACE);,,$$(subst $$(SPACE)=>$$(SPACE),"$$(RIGHT_PAREN)$$(RIGHT_PAREN) \
        { include$$(LEFT_PAREN)",$$(subst $$(SPACE);$$(SPACE),"$$(RIGHT_PAREN) } \
        else if $$(LEFT_PAREN)matches$$(LEFT_PAREN)",$$(strip $$($1_INCLUDES)))))
    $1_INCLUDES_COMMAND_LINE := $(NAWK) '$$($1_INCLUDES_HEADER_AWK) \
        { if (matches("$$($1_INCLUDES_PARTIAL_AWK)") } else print }'
  else
    # We don't have any includes, just pipe the file through cat.
    $1_INCLUDES_COMMAND_LINE := $(CAT)
  endif

  $1_VARDEPS := $$($1_INCLUDES_COMMAND_LINE) $$($1_REPLACEMENTS_COMMAND_LINE)
  $1_VARDEPS_FILE := $$(call DependOnVariable, $1_VARDEPS)

  # Reset target list before populating it
  $1 :=

  ifneq ($$($1_OUTPUT_FILE),)
    ifneq ($$(words $$($1_SOURCE_FILES)), 1)
      $$(error Cannot use OUTPUT_FILE for more than one source file (in $1))
    endif

    # Note that $1 is space sensitive and must disobey whitespace rules
    $$(eval $$(call SetupSingleTextFileForProcessing,$1, $$($1_SOURCE_FILES), \
        $$(patsubst %/, %, $$(dir $$($1_OUTPUT_FILE))), $$(notdir $$($1_OUTPUT_FILE))))
  else
    ifeq ($$($1_OUTPUT_DIR),)
      $$(error Neither OUTPUT_FILE nor OUTPUT_DIR was specified (in $1))
    endif

    # Now call add_native_source for each source file we are going to process.
    ifeq ($$($1_SOURCE_BASE_DIR),)
      # With no base dir specified, put all files in target dir, flattening any
      # hierarchies. Note that $1 is space sensitive and must disobey whitespace
      # rules.
      $$(foreach src, $$($1_SOURCE_FILES), \
          $$(eval $$(call SetupSingleTextFileForProcessing,$1, $$(src), \
              $$(patsubst %/, %, $$($1_OUTPUT_DIR)), $$(notdir $$(src)))))
    else
      # With a base dir, extract the relative portion of the path. Note that $1
      # is space sensitive and must disobey whitespace rules, and so is the
      # arguments to patsubst.
      $$(foreach src, $$($1_SOURCE_FILES), \
          $$(eval $$(call SetupSingleTextFileForProcessing,$1, $$(src), \
              $$(patsubst %/, %, $$($1_OUTPUT_DIR)), \
              $$(patsubst $$($1_SOURCE_BASE_DIR)/%,%,$$(src)))))
    endif
  endif
endef