From a20eb61c2fdd027a89acd249eea4f452e4accfb8 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?H=C3=A5kan=20Mattsson?= Date: Mon, 1 Mar 2010 19:53:48 +0100 Subject: Add function zip:foldl/3 to iterate over zip archives This is the public interface of prim_zip:open/3, which has been used in earlier releases by both erl_prim_loader and escript. The new function can be used as a replacement for the undocumented function escript:foldl/3 that is likely to be removed without further notice. The error handling of prim_zip:open/3 (and prim_zip:foldl/3) has been improved in order to better suite a public interface. For example it could happen that a file or a zlib port could be left open in some errors cases. The documentation of the FileSpec parameter to zip:create/3 has been updated to show that file info can be explicitly specified. A FileSpec may contain {Filename, binary(), #file_info{}} elements. The function zip:create/3 was already prepared to partly support this, but now after a few minor fixes it is fully supported. --- lib/stdlib/doc/src/zip.xml | 90 +++++++++++++++++++++++++++++++++++-------- lib/stdlib/src/zip.erl | 80 ++++++++++++++++++++++++++------------ lib/stdlib/test/zip_SUITE.erl | 78 ++++++++++++++++++++++++++----------- 3 files changed, 186 insertions(+), 62 deletions(-) (limited to 'lib') diff --git a/lib/stdlib/doc/src/zip.xml b/lib/stdlib/doc/src/zip.xml index af407f1925..4d98a20206 100644 --- a/lib/stdlib/doc/src/zip.xml +++ b/lib/stdlib/doc/src/zip.xml @@ -42,16 +42,18 @@

By convention, the name of a zip file should end in ".zip". To abide to the convention, you'll need to add ".zip" yourself to the name.

-

Zip archives are created with the - zip/2 or the +

Zip archives are created with the + zip/2 or the zip/3 function. (They are also available as create, to resemble the erl_tar module.)

-

To extract files from a zip archive, use the - unzip/1 or the +

To extract files from a zip archive, use the + unzip/1 or the unzip/2 function. (They are also available as extract.)

-

To return a list of the files in a zip archive, use the +

To fold a function over all files in a zip archive, use the + foldl_3.

+

To return a list of the files in a zip archive, use the list_dir/1 or the list_dir/2 function. (They are also available as table.)

@@ -132,7 +134,7 @@ zip_file() Name = filename() FileList = [FileSpec] - FileSpec = filename() | {filename(), binary()} + FileSpec = filename() | {filename(), binary()} | {filename(), binary(), #file_info{}} Options = [Option] Option = memory | cooked | verbose | {comment, Comment} | {cwd, CWD} | {compress, What} | {uncompress, What} What = all | [Extension] | {add, [Extension]} | {del, [Extension]} @@ -212,16 +214,16 @@ zip_file() all

means that all files will be compressed (as long - as they pass the uncompress condition).

 + as they pass the uncompress condition).

[Extension]

means that only files with exactly these extensions - will be compressed.

 + will be compressed.

{add,[Extension]}

adds these extensions to the list of compress - extensions.

 + extensions.

{del,[Extension]}

deletes these extensions from the list of compress - extensions.

 + extensions.

 {uncompress, What} @@ -231,16 +233,16 @@ zip_file() The following values of What are allowed:

all -

means that no files will be compressed.

 +

means that no files will be compressed.

 [Extension]

means that files with these extensions will be - uncompressed.

 + uncompressed.

{add,[Extension]}

adds these extensions to the list of uncompress - extensions.

 + extensions.

{del,[Extension]}

deletes these extensions from the list of uncompress - extensions.

 + extensions.

 @@ -283,7 +285,7 @@ zip_file() the unzip/2 function will only extract the files whose names are included in FileList. The full paths, including the names of all sub directories within - the zip archive, must be specified.

+ the zip archive, must be specified.

cooked @@ -326,6 +328,64 @@ zip_file() + + foldl(Fun, Acc0, Archive) -> {ok, Acc1} | {error, Reason} + Fold a function over all files in a zip archive + + Fun = fun(FileInArchive, GetInfo, GetBin, AccIn) -> AccOut + FileInArchive = filename() + GetInfo = fun() -> #file_info{} + GetBin = fun() -> binary() + Acc0 = Acc1 = AccIn = AccOut = term() + Archive = filename() | {filename(), binary()} + + +

The foldl/3 function + calls Fun(FileInArchive, GetInfo, GetBin, AccIn) on + successive files in the Archive, starting with AccIn + == Acc0. FileInArchive is the name that the file + has in the archive. GetInfo is a fun that returns info + about the the file. GetBin returns the contents of the + file. Both GetInfo and GetBin must be called + within the Fun. Their behavior is undefined if they are + called outside the context of the Fun. The Fun + must return a new accumulator which is passed to the next + call. foldl/3 returns the final value of the + accumulator. Acc0 is returned if the archive is + empty. It is not necessary to iterate over all files in the + archive. The iteration may be ended prematurely in a + controlled manner by throwing an exception.

+ +

For example:

+ 
+> Name = "dummy.zip".
+"dummy.zip"
+> {ok, {Name, Bin}} = zip:create(Name, [{"foo", <<"FOO">>}, {"bar", <<"BAR">>}], [memory]).
+{ok,{"dummy.zip",
+     <<80,75,3,4,20,0,0,0,0,0,74,152,97,60,171,39,212,26,3,0,
+       0,0,3,0,0,...>>}}
+> {ok, FileSpec} = zip:foldl(fun(N, I, B, Acc) -> [{N, B(), I()} | Acc] end, [], {Name, Bin}).
+{ok,[{"bar",<<"BAR">>,
+      {file_info,3,regular,read_write,
+                 {{2010,3,1},{19,2,10}},
+                 {{2010,3,1},{19,2,10}},
+                 {{2010,3,1},{19,2,10}},
+                 54,1,0,0,0,0,0}},
+     {"foo",<<"FOO">>,
+      {file_info,3,regular,read_write,
+                 {{2010,3,1},{19,2,10}},
+                 {{2010,3,1},{19,2,10}},
+                 {{2010,3,1},{19,2,10}},
+                 54,1,0,0,0,0,0}}]}
+> {ok, {Name, Bin}} = zip:create(Name, lists:reverse(FileSpec), [memory]).
+{ok,{"dummy.zip",
+     <<80,75,3,4,20,0,0,0,0,0,74,152,97,60,171,39,212,26,3,0,
+       0,0,3,0,0,...>>}}
+> catch zip:foldl(fun("foo", _, B, _) -> throw(B()); (_, _, _, Acc) -> Acc end, [], {Name, Bin}). 
+<<"FOO">>
+
+ + list_dir(Archive) -> RetValue list_dir(Archive, Options) diff --git a/lib/stdlib/src/zip.erl b/lib/stdlib/src/zip.erl index e76d588cb5..d41aeefa59 100644 --- a/lib/stdlib/src/zip.erl +++ b/lib/stdlib/src/zip.erl @@ -20,7 +20,7 @@ %% Basic api -export([unzip/1, unzip/2, extract/1, extract/2, - zip/2, zip/3, create/2, create/3, + zip/2, zip/3, create/2, create/3, foldl/3, list_dir/1, list_dir/2, table/1, table/2, t/1, tt/1]). @@ -38,7 +38,7 @@ zip_t/1, zip_tt/1, zip_list_dir/1, zip_list_dir/2, zip_close/1]). - + %% just for debugging zip server, not documented, not tested, not to be used -export([zip_get_state/1]). @@ -82,7 +82,7 @@ -record(openzip_opts, { output, % output object (fun) open_opts, % file:open options - cwd % directory to relate paths to + cwd % directory to relate paths to }). % openzip record, state for an open zip-file @@ -93,10 +93,10 @@ input, % archive io object (fun) output, % output io object (fun) zlib, % handle to open zlib - cwd % directory to relate paths to + cwd % directory to relate paths to }). -% Things that I would like to add to the public record #zip_file, +% Things that I would like to add to the public record #zip_file, % but can't as it would make things fail at upgrade. % Instead we use {#zip_file,#zip_file_extra} internally. -record(zip_file_extra, { @@ -278,7 +278,7 @@ file_name_search(Name,Files) -> [ZFile|_] -> ZFile; [] -> false end. - + %% %% add a file to an open archive %% openzip_add(File, OpenZip) -> %% case ?CATCH do_openzip_add(File, OpenZip) of @@ -344,6 +344,25 @@ do_unzip(F, Options) -> Input(close, In1), {ok, Files}. +%% Iterate over all files in a zip archive +foldl(Fun, Acc0, Archive) when is_function(Fun, 4) -> + ZipFun = + fun({Name, GetInfo, GetBin}, A) -> + A2 = Fun(Name, GetInfo, GetBin, A), + {true, false, A2} + end, + case prim_zip:open(ZipFun, Acc0, Archive) of + {ok, PrimZip, Acc1} -> + ok = prim_zip:close(PrimZip), + {ok, Acc1}; + {error, bad_eocd} -> + {error, "Not an archive file"}; + {error, Reason} -> + {error, Reason} + end; +foldl(_,_, _) -> + {error, einval}. + %% Create zip archive name F from Files or binaries %% %% Accepted options: @@ -383,7 +402,7 @@ list_dir(F, Options) -> do_list_dir(F, Options) -> Opts = get_list_dir_options(F, Options), - #list_dir_opts{input = Input, open_opts = OpO, + #list_dir_opts{input = Input, open_opts = OpO, raw_iterator = RawIterator} = Opts, In0 = Input({open, F, OpO}, []), {Info, In1} = get_central_dir(In0, RawIterator, Input), @@ -417,7 +436,7 @@ tt(F) when is_record(F, openzip) -> openzip_tt(F); tt(F) -> t(F, fun raw_long_print_info_etc/5). -%% option utils +%% option utils get_unzip_opt([], Opts) -> Opts; get_unzip_opt([verbose | Rest], Opts) -> @@ -470,7 +489,7 @@ get_zip_opt([{cwd, CWD} | Rest], Opts) -> get_zip_opt([{comment, C} | Rest], Opts) -> get_zip_opt(Rest, Opts#zip_opts{comment = C}); get_zip_opt([{compress, Which} = O| Rest], Opts) -> - Which2 = + Which2 = case Which of all -> all; @@ -485,7 +504,7 @@ get_zip_opt([{compress, Which} = O| Rest], Opts) -> end, get_zip_opt(Rest, Opts#zip_opts{compress = Which2}); get_zip_opt([{uncompress, Which} = O| Rest], Opts) -> - Which2 = + Which2 = case Which of all -> all; @@ -560,16 +579,24 @@ get_openzip_options(Options) -> get_input(F) when is_binary(F) -> fun binary_io/2; get_input(F) when is_list(F) -> - fun file_io/2. + fun file_io/2; +get_input(_) -> + throw(einval). get_zip_input({F, B}) when is_binary(B), is_list(F) -> fun binary_io/2; +get_zip_input({F, B, #file_info{}}) when is_binary(B), is_list(F) -> + fun binary_io/2; +get_zip_input({F, #file_info{}, B}) when is_binary(B), is_list(F) -> + fun binary_io/2; get_zip_input(F) when is_list(F) -> fun file_io/2; get_zip_input({files, []}) -> fun binary_io/2; get_zip_input({files, [File | _]}) -> - get_zip_input(File). + get_zip_input(File); +get_zip_input(_) -> + throw(einval). get_list_dir_options(F, Options) -> Opts = #list_dir_opts{raw_iterator = fun raw_file_info_public/5, @@ -620,6 +647,8 @@ put_eocd(N, Pos, Sz, Comment, Output, Out0) -> get_filename({Name, _}, Type) -> get_filename(Name, Type); +get_filename({Name, _, _}, Type) -> + get_filename(Name, Type); get_filename(Name, regular) -> Name; get_filename(Name, directory) -> @@ -895,7 +924,7 @@ local_file_header_to_bin( CompSize:32/little, UncompSize:32/little, FileNameLength:16/little, - ExtraFieldLength:16/little>>. + ExtraFieldLength:16/little>>. eocd_to_bin(#eocd{disk_num = DiskNum, start_disk_num = StartDiskNum, @@ -912,7 +941,7 @@ eocd_to_bin(#eocd{disk_num = DiskNum, Offset:32/little, ZipCommentLength:16/little>>. -%% put together a local file header +%% put together a local file header local_file_header_from_info_method_name(#file_info{mtime = MTime}, UncompSize, CompMethod, Name) -> @@ -939,7 +968,7 @@ server_loop(OpenZip) -> server_loop(NewOpenZip); Error -> From ! {self(), Error} - end; + end; {From, close} -> From ! {self(), openzip_close(OpenZip)}; {From, get} -> @@ -1024,7 +1053,7 @@ lists_foreach(F, [Hd|Tl]) -> F(Hd), lists_foreach(F, Tl). -%% option utils +%% option utils get_openzip_opt([], Opts) -> Opts; get_openzip_opt([cooked | Rest], #openzip_opts{open_opts = OO} = Opts) -> @@ -1121,7 +1150,7 @@ raw_file_info_public(CD, FileName, FileComment, BExtraField, Acc0) -> Other -> Other end, [H2|T]. - + %% make a file_info from a central directory header cd_file_header_to_file_info(FileName, @@ -1213,8 +1242,8 @@ get_z_file(In0, Z, Input, Output, OpO, FB, CWD, {ZipFile,Extra}) -> {dir, In3}; _ -> %% FileInfo = local_file_header_to_file_info(LH) - %%{Out, In4, CRC, UncompSize} = - {Out, In4, CRC, _UncompSize} = + %%{Out, In4, CRC, UncompSize} = + {Out, In4, CRC, _UncompSize} = get_z_data(CompMethod, In3, FileName1, CompSize, Input, Output, OpO, Z), In5 = skip_z_data_descriptor(GPFlag, Input, In4), @@ -1280,7 +1309,7 @@ get_z_data_loop(CompSize, UncompSize, In0, Out0, Input, Output, Z) -> Out1 = Output({write, Uncompressed}, Out0), get_z_data_loop(CompSize-N, UncompSize + iolist_size(Uncompressed), In1, Out1, Input, Output, Z) - end. + end. %% skip data descriptor if any @@ -1298,7 +1327,7 @@ dos_date_time_to_datetime(DosDate, DosTime) -> <> = <>, <> = <>, {{YearFrom1980+1980, Month, Day}, - {Hour, Min, Sec}}. + {Hour, Min, Sec}}. dos_date_time_from_datetime({{Year, Month, Day}, {Hour, Min, Sec}}) -> YearFrom1980 = Year-1980, @@ -1319,7 +1348,6 @@ unix_extra_field_and_var_from_bin(< throw(bad_unix_extra_field). - %% A pwrite-like function for iolists (used by memory-option) @@ -1478,6 +1506,8 @@ local_file_header_from_bin(_) -> %% io functions binary_io({file_info, {_Filename, _B, #file_info{} = FI}}, _A) -> FI; +binary_io({file_info, {_Filename, #file_info{} = FI, _B}}, _A) -> + FI; binary_io({file_info, {_Filename, B}}, A) -> binary_io({file_info, B}, A); binary_io({file_info, B}, _) -> @@ -1493,9 +1523,11 @@ binary_io({file_info, B}, _) -> links = 1, major_device = 0, minor_device = 0, inode = 0, uid = 0, gid = 0}; -binary_io({open, {_Filename, B, _FI}, _Opts}, _) -> +binary_io({open, {_Filename, B, _FI}, _Opts}, _) when is_binary(B) -> + {0, B}; +binary_io({open, {_Filename, _FI, B}, _Opts}, _) when is_binary(B) -> {0, B}; -binary_io({open, {_Filename, B}, _Opts}, _) -> +binary_io({open, {_Filename, B}, _Opts}, _) when is_binary(B) -> {0, B}; binary_io({open, B, _Opts}, _) when is_binary(B) -> {0, B}; diff --git a/lib/stdlib/test/zip_SUITE.erl b/lib/stdlib/test/zip_SUITE.erl index 12ca655000..48b14396c1 100644 --- a/lib/stdlib/test/zip_SUITE.erl +++ b/lib/stdlib/test/zip_SUITE.erl @@ -18,12 +18,13 @@ %% -module(zip_SUITE). --export([all/1, borderline/1, atomic/1, +-export([all/1, borderline/1, atomic/1, bad_zip/1, unzip_from_binary/1, unzip_to_binary/1, zip_to_binary/1, unzip_options/1, zip_options/1, list_dir_options/1, aliases/1, openzip_api/1, zip_api/1, unzip_jar/1, - compress_control/1]). + compress_control/1, + foldl/1]). -include("test_server.hrl"). -include("test_server_line.hrl"). @@ -35,7 +36,8 @@ all(suite) -> [borderline, atomic, bad_zip, zip_to_binary, unzip_options, zip_options, list_dir_options, aliases, openzip_api, zip_api, unzip_jar, - compress_control]. + compress_control, + foldl]. borderline(doc) -> ["Test creating, listing and extracting one file from an archive " @@ -110,17 +112,17 @@ get_data(Port, Expect) -> {Port, {data, Bytes}} -> get_data(Port, match_output(Bytes, Expect, Port)); {Port, eof} -> - Port ! {self(), close}, + Port ! {self(), close}, receive {Port, closed} -> true - end, + end, receive - {'EXIT', Port, _} -> + {'EXIT', Port, _} -> ok after 1 -> % force context switch ok - end, + end, match_output(eof, Expect, Port) end. @@ -290,7 +292,7 @@ unzip_options(Config) when is_list(Config) -> DataDir = ?config(data_dir, Config), PrivDir = ?config(priv_dir, Config), Long = filename:join(DataDir, "abc.zip"), - + %% create a temp directory Subdir = filename:join(PrivDir, "t"), ok = file:make_dir(Subdir), @@ -303,7 +305,7 @@ unzip_options(Config) when is_list(Config) -> %% Verify. ?line true = (length(FList) =:= length(RetList)), - ?line lists:foreach(fun(F)-> {ok,B} = file:read_file(filename:join(DataDir, F)), + ?line lists:foreach(fun(F)-> {ok,B} = file:read_file(filename:join(DataDir, F)), {ok,B} = file:read_file(filename:join(Subdir, F)) end, FList), ?line lists:foreach(fun(F)-> ok = file:delete(F) end, @@ -321,7 +323,7 @@ unzip_jar(Config) when is_list(Config) -> DataDir = ?config(data_dir, Config), PrivDir = ?config(priv_dir, Config), JarFile = filename:join(DataDir, "test.jar"), - + %% create a temp directory Subdir = filename:join(PrivDir, "jartest"), ok = file:make_dir(Subdir), @@ -479,7 +481,7 @@ unzip_to_binary(Config) when is_list(Config) -> DataDir = ?config(data_dir, Config), PrivDir = ?config(priv_dir, Config), - delete_all_in(PrivDir), + delete_all_in(PrivDir), file:set_cwd(PrivDir), Long = filename:join(DataDir, "abc.zip"), @@ -595,17 +597,17 @@ do_delete_files([],Cnt) -> Cnt; do_delete_files([Item|Rest], Cnt) -> case file:delete(Item) of - ok -> + ok -> DelCnt = 1; {error,eperm} -> file:change_mode(Item, 8#777), DelCnt = delete_files(filelib:wildcard(filename:join(Item, "*"))), - file:del_dir(Item); + file:del_dir(Item); {error,eacces} -> %% We'll see about that! file:change_mode(Item, 8#777), case file:delete(Item) of - ok -> + ok -> DelCnt = 1; {error,_} -> erlang:yield(), @@ -643,22 +645,22 @@ compress_control(Config) when is_list(Config) -> ], test_compress_control(Dir, - Files, + Files, [{compress, []}], []), test_compress_control(Dir, - Files, + Files, [{uncompress, all}], []), test_compress_control(Dir, - Files, + Files, [{uncompress, []}], [".txt", ".exe", ".zip", ".lzh", ".arj"]), test_compress_control(Dir, - Files, + Files, [], [".txt", ".exe"]), @@ -686,7 +688,7 @@ test_compress_control(Dir, Files, ZipOptions, Expected) -> create_files(Files), {ok, Zip} = zip:create(Zip, [Dir], ZipOptions), - + {ok, OpenZip} = zip:openzip_open(Zip, [memory]), {ok,[#zip_comment{comment = ""} | ZipList]} = zip:openzip_list_dir(OpenZip), io:format("compress_control: -> ~p -> ~p\n -> ~pn", [Expected, ZipOptions, ZipList]), @@ -698,19 +700,19 @@ test_compress_control(Dir, Files, ZipOptions, Expected) -> delete_files(lists:reverse(Names)), % Remove plain files before directories ok. - + verify_compression([{Name, Kind, _Filler} | Files], ZipList, OpenZip, ZipOptions, Expected) -> {Name2, BinSz} = case Kind of - dir -> + dir -> {Name ++ "/", 0}; - _ -> + _ -> {ok, {Name, Bin}} = zip:openzip_get(Name, OpenZip), {Name, size(Bin)} end, {Name2, {value, ZipFile}} = {Name2, lists:keysearch(Name2, #zip_file.name, ZipList)}, #zip_file{info = #file_info{size = InfoSz, type = InfoType}, comp_size = InfoCompSz} = ZipFile, - + Ext = filename:extension(Name), IsComp = is_compressed(Ext, Kind, ZipOptions), ExpComp = lists:member(Ext, Expected), @@ -757,3 +759,33 @@ extensions([H | T], Old) -> extensions([], Old) -> Old. +foldl(Config) -> + PrivDir = ?config(priv_dir, Config), + File = filename:join([PrivDir, "foldl.zip"]), + + FooBin = <<"FOO">>, + BarBin = <<"BAR">>, + Files = [{"foo", FooBin}, {"bar", BarBin}], + ?line {ok, {File, Bin}} = zip:create(File, Files, [memory]), + ZipFun = fun(N, I, B, Acc) -> [{N, B(), I()} | Acc] end, + ?line {ok, FileSpec} = zip:foldl(ZipFun, [], {File, Bin}), + ?line [{"bar", BarBin, #file_info{}}, {"foo", FooBin, #file_info{}}] = FileSpec, + ?line {ok, {File, Bin}} = zip:create(File, lists:reverse(FileSpec), [memory]), + ?line {foo_bin, FooBin} = + try + zip:foldl(fun("foo", _, B, _) -> throw(B()); (_, _, _, Acc) -> Acc end, [], {File, Bin}) + catch + throw:FooBin -> + {foo_bin, FooBin} + end, + ?line ok = file:write_file(File, Bin), + ?line {ok, FileSpec} = zip:foldl(ZipFun, [], File), + + ?line {error, einval} = zip:foldl(fun() -> ok end, [], File), + ?line {error, einval} = zip:foldl(ZipFun, [], 42), + ?line {error, einval} = zip:foldl(ZipFun, [], {File, 42}), + + ?line ok = file:delete(File), + ?line {error, enoent} = zip:foldl(ZipFun, [], File), + + ok. -- cgit v1.2.3