[developers] Compiling FFTB on Ubuntu 19.10

Woodley Packard sweaglesw at sweaglesw.org
Mon Mar 30 23:11:35 CEST 2020


Hi Alex,

It looks like compiling the library succeeded but the test app failed, most likely just because the library is not yet installed.  Please install the libraries (make install, or however you prefer) and that should allow the test app to build.

-Woodley

> On Mar 30, 2020, at 12:47 PM, Alexandre Rademaker <arademaker at gmail.com> wrote:
> 
> 
> Hi Woodley,
> 
> I had to reinstall my machine and now I am trying to recompile all the tools. I gave up for compiling them for MacOS. That would be great for me, but in the MacOS I haven’t passed from the first step below. So I am compiling everything in a docker running Ubuntu 19.10.
> 
> My goal is to have FFTB running again, I can't use the acetools binaries you provided because it seems that http://sweaglesw.org/svn/treebank/trunk/web.c still don’t have the changed you made during a conversation in Cambridge:
> 
> Line 1290:
> 
>  addr.sin_addr.s_addr = 0; // inet_addr("127.0.0.1”);
> 
> Without that change, running FFTB inside the docker is not easy. We need a proxy server for redirecting the ports (as documented in http://moin.delph-in.net/FftbTop#FFTB_on_remote_machine), but with that change, we don’t need to proxy and can use the docker native way to redirect internal to external ports.
> 
> I have tried to follow the steps that worked for me last time:
> 
> 1. install liba
> 2. install repp-0.2.2
> 3. install libace
> 4. Install libtsdb
> 
> I have success for 1-3, but in step 4 I got an error. The error was caused by
> 
> /usr/bin/ld: cannot find -ltsdb
> 
> This is a little bit strange because it seems that during the compilation of libtsdb it is looking for this same library… am I right?
> 
> The complete trace is below. I didn’t see this error before. Can you help me? I am blocked by this error...
> 
> 
> $ make
> cc -fPIC -shared -g -O2   -c -o tsdb.o tsdb.c
> tsdb.c: In function ‘tsdb_free_profile’:
> tsdb.c:47:5: warning: implicit declaration of function ‘hash_free_nokeys’ [-Wimplicit-function-declaration]
>   47 |     hash_free_nokeys(r->fields[j].hash);
>      |     ^~~~~~~~~~~~~~~~
> tsdb.c: In function ‘tsdb_write_relation’:
> tsdb.c:258:2: warning: implicit declaration of function ‘unlink’ [-Wimplicit-function-declaration]
>  258 |  unlink(fname_bk);
>      |  ^~~~~~
> cc -fPIC -shared -g -O2   -c -o relations.o relations.c
> gcc -fPIC -shared -g -O2 -fvisibility=hidden -c hash.c -o hash.o
> gcc -fPIC -shared -g -O2 tsdb.o relations.o hash.o -shared -o libtsdb.so
> rm -f libtsdb.a
> ar cru libtsdb.a tsdb.o relations.o
> ar: `u' modifier ignored since `D' is the default (see `U')
> gcc -g -O2 -L. test.c -ltsdb -o test -Wl,-rpath -Wl,`pwd` -lace -ldl -la
> test.c:74:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>   74 | print_tree_with_edge_id(struct tree *t, int indent, int *edgemap)
>      | ^~~~~~~~~~~~~~~~~~~~~~~
> test.c:122:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>  122 | record_eq_edges(int x_eid, int y_eid)
>      | ^~~~~~~~~~~~~~~
> test.c: In function ‘report_missing_edges’:
> test.c:165:3: warning: implicit declaration of function ‘print_tree’; did you mean ‘print_mrs’? [-Wimplicit-function-declaration]
>  165 |   print_tree(t, 2);
>      |   ^~~~~~~~~~
>      |   print_mrs
> test.c: At top level:
> test.c:174:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>  174 | fidget(struct tree *t)
>      | ^~~~~~
> test.c:190:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>  190 | compare_tree_lists(char *iid, struct result *rx, int nx, struct result *ry, int ny, int detail, char *errx, char *erry)
>      | ^~~~~~~~~~~~~~~~~~
> test.c: In function ‘compare_tree_lists’:
> test.c:277:2: warning: implicit declaration of function ‘hash_free’; did you mean ‘hash_find’? [-Wimplicit-function-declaration]
>  277 |  hash_free(hx);
>      |  ^~~~~~~~~
>      |  hash_find
> test.c: In function ‘tree_to_mrs’:
> test.c:381:2: warning: implicit declaration of function ‘clear_mrs’; did you mean ‘read_mrs’? [-Wimplicit-function-declaration]
>  381 |  clear_mrs();
>      |  ^~~~~~~~~
>      |  read_mrs
> test.c: At top level:
> test.c:421:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>  421 | compare_surface_lists(char *iid, char *i_input, struct result *rx, int nx, struct result *ry, int ny, struct tree *gold_tree, struct mrs *gold_mrs, int detail)
>      | ^~~~~~~~~~~~~~~~~~~~~
> test.c:507:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>  507 | usage(char *prog)
>      | ^~~~~
> test.c:585:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>  585 | main(int argc, char *argv[])
>      | ^~~~
> test.c: In function ‘main’:
> test.c:599:2: warning: implicit declaration of function ‘ace_load_grammar’ [-Wimplicit-function-declaration]
>  599 |  ace_load_grammar("/home/sweaglesw/cdev/ace-regression/comparison.dat");
>      |  ^~~~~~~~~~~~~~~~
> gcc -g -O2 art.c -lace -ltsdb -lrepp -la -o art -lutil
> art.c:65:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>   65 | usage(char *myname, int status)
>      | ^~~~~
> art.c:87:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>   87 | main(int argc, char *argv[])
>      | ^~~~
> art.c: In function ‘main’:
> art.c:156:13: warning: implicit declaration of function ‘forkpty’; did you mean ‘fork’? [-Wimplicit-function-declaration]
>  156 |   pid_t p = forkpty(&arbiter_fd, NULL, NULL, NULL);
>      |             ^~~~~~~
>      |             fork
> art.c:301:16: warning: implicit declaration of function ‘read_result’; did you mean ‘record_result’? [-Wimplicit-function-declaration]
>  301 |   int status = read_result(parse_id, run_id, i_id, i_input);
>      |                ^~~~~~~~~~~
>      |                record_result
> art.c: At top level:
> art.c:560:1: warning: return type defaults to ‘int’ [-Wimplicit-int]
>  560 | write_tuple(FILE *f, char **tuple, struct relation *r)
>      | ^~~~~~~~~~~
> /usr/bin/ld: cannot find -ltsdb
> collect2: error: ld returned 1 exit status
> make: *** [Makefile:51: art] Error 1
> 
> 
> Best,
> Alexandre
> 
> 
> 




More information about the developers mailing list