Improve the logic, and make it more modular.

This commit is contained in:
j2blake 2011-12-01 22:35:09 +00:00
parent d0d051e072
commit 7b5660f82a

View file

@ -15,7 +15,109 @@ require 'date'
require 'fileutils' require 'fileutils'
require 'property_file_reader' require 'property_file_reader'
NAMESPACES_FILENAME = 'file_storage_namespaces.properties'
# ------------------------------------------------------------------------------------
# FileInfoFile class
# ------------------------------------------------------------------------------------
class FileInfoFile
attr_reader :default_namespace
attr_reader :data
def parse_info_line(line)
# Lines are in this form: "URI","filename"
match = line.match(/^"([^"]*)","([^"]*)"$/)
raise "Can't parse this line: '#{line}'" if !match
return match.captures[0], match.captures[1]
end
def parse_default_namespace(uri)
match = /^(.*)individual/.match(uri)
raise "Can't find default namespace: '#{uri}'" if match == nil
"#{match.captures[0]}individual/"
end
public
def initialize(filepath)
@data = []
File.open(filepath) do |f|
f.each() do |line|
@data.push(parse_info_line(line))
end
end
puts "parsed #{@data.length} lines."
@default_namespace = parse_default_namespace(@data[0][0])
puts "default namespace is '#{@default_namespace}'"
end
end
# ------------------------------------------------------------------------------------
# NamespacesFile class
#
# Read, query, modify and write the namespace-prefixes file.
# ------------------------------------------------------------------------------------
class NamespacesFile
NAMESPACES_FILENAME = 'file_storage_namespaces.properties'
public
def initialize(uploads_directory, scan_only)
@uploads_directory = uploads_directory
@scan_only = scan_only
Dir.chdir(@uploads_directory) do |dir|
@namespaces = {}
if File.file?(NAMESPACES_FILENAME)
@namespaces = PropertyFileReader.read(NAMESPACES_FILENAME)
@namespaces.delete("properties_file_path")
end
end
end
def add_namespace(namespace)
if @namespaces.has_value?(namespace)
puts "found prefix for #{namespace}"
return
end
'abcdefghijklmnopqrstuvwxyz'.split("").each do |this_char|
if (!@namespaces.has_key?(this_char))
@namespaces[this_char] = namespace
puts "assigned prefix = '#{this_char}'"
return
end
end
raise "all prefixes are used!"
end
def prefix(namespace)
@namespaces.each() do | key, value |
return key if value == namespace
end
raise "no prefix for '#{namespace}'"
end
def write()
if @scan_only
puts "Scan-only: not writing namespaces file"
else
Dir.chdir(@uploads_directory) do |dir|
File.open(NAMESPACES_FILENAME, "w") do |f|
@namespaces.each do |prefix, namespace|
f.puts "#{prefix} = #{namespace}"
end
end
end
end
end
end
# ------------------------------------------------------------------------------------
# Main class - UploadFileFaker
# ------------------------------------------------------------------------------------
class UploadFileFaker class UploadFileFaker
# #
@ -38,103 +140,18 @@ class UploadFileFaker
end end
# #
# Look in the file_info file and find the default namespace. Lines are in # Check each location that should contain an image, and if we're not just
# this form: # scanning, put one there.
# Uv::http://vivo.scripps.edu/individual/n2938: Lv:0::_main_image_testtubes.jpg:
# #
# The first line may be a header line, so use the second one. def create_image_files_where_needed()
# @file_info.data.each do |line|
def find_default_namespace() uri, filename = line
second_line = IO.readlines(@file_info_file)[1] process_file_info(uri, filename)
match = /Uv::(.*)individual/.match(second_line)
raise "Can't find default namespace: '#{second_line}'" if match == nil
@default_namespace = "#{match.captures[0]}individual/"
puts "default namespace is '#{@default_namespace}'"
end
def adjust_namespaces_file()
namespaces = read_namespaces_file()
namespaces = add_namespace_if_needed(namespaces)
write_namespaces_file(namespaces)
namespaces.each() do |key, value|
if value == @default_namespace
@prefix = key
puts "prefix is #{@prefix}"
break
end
end end
end end
def read_namespaces_file() def process_file_info(uri, filename)
Dir.chdir(@uploads_directory) do |dir| full_path = figure_full_path(uri, filename)
namespaces = {}
if File.file?(NAMESPACES_FILENAME)
namespaces = PropertyFileReader.read(NAMESPACES_FILENAME)
namespaces.delete("properties_file_path")
puts "namespaces is already set to '#{namespaces}'"
end
return namespaces
end
end
def add_namespace_if_needed(namespaces)
if namespaces.has_value?(@default_namespace)
puts "found prefix"
else
@prefix = ''
'abcdefghijklmnopqrstuvwxyz'.chars() do |this_char|
if (!namespaces.has_key?(this_char))
namespaces[@prefix] = @default_namespace
puts "assigned prefix = '#{@this_char}'"
break
end
end
end
return namespaces
end
def write_namespaces_file(namespaces)
if @scan_only
puts "Scan-only: not writing namespaces file"
else
Dir.chdir(@uploads_directory) do |dir|
File.open(NAMESPACES_FILENAME, "w") do |f|
namespaces.each do |prefix, namespace|
f.puts "#{prefix} = #{namespace}"
end
end
end
end
end
#
# Second pass -- figure out the location for each file. If the file
# doesn't exist, copy the template to that location.
#
def second_pass()
File.open(@file_info_file) do |f|
f.each() do |line|
next if header_line?(line.chomp)
process_file_info_line(line.chomp)
end
end
end
def header_line?(line)
return false unless line.match("^bytestreamUri")
puts "Skipping header line: '#{line}'"
return true
end
def process_file_info_line(line)
uri, filename = parse_info_line(line)
puts "URI is '#{uri}'"
puts "Filename is '#{filename}'"
prefixed_uri = substitute_prefix_for_namespace(uri)
full_path = construct_full_path(prefixed_uri, filename)
puts "Full path is '#{full_path}'"
if File.file?(full_path) if File.file?(full_path)
puts "File already exists at: '#{full_path}'" puts "File already exists at: '#{full_path}'"
@ -147,16 +164,14 @@ class UploadFileFaker
end end
end end
def parse_info_line(line) def figure_full_path(uri, filename)
# Lines are in this form: Uv::URI: Lv:0::FILENAME: prefixed_uri = substitute_prefix_for_namespace(uri)
match = line.match(/Uv::(.*):\s+Lv:0::(.*):/) construct_full_path(prefixed_uri, filename)
raise "Can't parse this line: '#{line}'" if !match
return match.captures[0], match.captures[1]
end end
def substitute_prefix_for_namespace(uri) def substitute_prefix_for_namespace(uri)
if uri[0, @default_namespace.length] == @default_namespace if uri[0, @namespace.length] == @namespace
return uri.sub(@default_namespace, "#{@prefix}~") uri.sub(@namespace, "#{@prefix}~")
else else
raise "Doesn't start with default namespace: '#{uri}'" raise "Doesn't start with default namespace: '#{uri}'"
end end
@ -167,13 +182,10 @@ class UploadFileFaker
0.step(prefixed_uri.size - 1, 3) do |i| 0.step(prefixed_uri.size - 1, 3) do |i|
path = "#{path}/#{prefixed_uri[i, 3]}" path = "#{path}/#{prefixed_uri[i, 3]}"
end end
puts "path is: '#{path}'" "#{@uploads_directory}/file_storage_root#{path}/#{filename}"
return "#{@uploads_directory}/file_storage_root#{path}/#{filename}"
end end
# ------------------------------------------------------------------------------------
public public
# ------------------------------------------------------------------------------------
# #
# Setup and get ready to process. # Setup and get ready to process.
@ -192,12 +204,18 @@ class UploadFileFaker
end end
# #
# Start the recursive scanning (and copying). # Start the scanning (and copying).
# #
def process() def process()
find_default_namespace() @file_info = FileInfoFile.new(@file_info_file)
adjust_namespaces_file() @namespace = @file_info.default_namespace
second_pass()
namespaces_file = NamespacesFile.new(@uploads_directory, @scan_only)
namespaces_file.add_namespace(@namespace)
namespaces_file.write()
@prefix = namespaces_file.prefix(@namespace)
create_image_files_where_needed()
end end
end end