Quellcodebibliothek Statistik Leitseite products/sources/formale Sprachen/C/LibreOffice/bin/   (Office von Apache Version 25.8.3.2©)  Datei vom 5.10.2025 mit Größe 10 kB image not shown  

Quelle  find-mergedlib-can-be-private.py   Sprache: Python

 
#!/usr/bin/python3
#
# Generate a custom linker script/map file for the --enabled-mergedlibs merged library
# which reduces the startup time and enables further optimisations with --enable-lto because 60% or more
# of the symbols become internal only.
#

import subprocess
import re

exported_symbols1 = set()
imported_symbols1 = set()
exported_symbols2 = set() # decoded
imported_symbols2 = set() # decoded


# find all our shared libs, note that all the libs that have been merged into the mergedlib library will
# not contain anything, those files are just empty
subprocess_find = subprocess.Popen("find ./instdir -name *.so && find ./workdir/LinkTarget/CppunitTest -name *.so",
    stdout=subprocess.PIPE, shell=True)
with subprocess_find.stdout as txt:
    for line in txt:
        sharedlib = line.strip()
        # look for exported symbols
        subprocess_nm = subprocess.Popen(b"nm -D " + sharedlib, stdout=subprocess.PIPE, shell=True)
        with subprocess_nm.stdout as txt2:
            # We are looking for lines something like:
            # 0000000000036ed0 T flash_component_getFactory
            line_regex = re.compile(r'^[0-9a-fA-F]+ T ')
            for line2_bytes in txt2:
                line2 = line2_bytes.strip().decode("utf-8")
                if line_regex.match(line2):
                    sym = line2.split(" ")[2].strip()
                    exported_symbols1.add(sym)
        subprocess_nm.terminate()
        # look for imported symbols
        subprocess_objdump = subprocess.Popen(b"objdump -T " + sharedlib, stdout=subprocess.PIPE, shell=True)
        with subprocess_objdump.stdout as txt2:
            # ignore some header bumpf
            txt2.readline()
            txt2.readline()
            txt2.readline()
            txt2.readline()
            # We are looking for lines something like:
            # 0000000000000000      DF *UND*  0000000000000000     _ZN16FilterConfigItem10WriteInt32ERKN3rtl8OUStringEi
            for line2_bytes in txt2:
                line2 = line2_bytes.strip().decode("utf-8")
                if "*UND*" not in line2:
                    continue
                tokens = line2.split(" ")
                sym = tokens[len(tokens)-1].strip()
                imported_symbols1.add(sym)
        subprocess_objdump.terminate()
subprocess_find.terminate()

# look for imported symbols in executables
subprocess_find = subprocess.Popen("find ./instdir -name *.bin", stdout=subprocess.PIPE, shell=True)
with subprocess_find.stdout as txt:
    for line in txt:
        executable = line.strip()
        # look for exported symbols
        subprocess_nm = subprocess.Popen(b"nm -D " + executable + b" | grep -w U", stdout=subprocess.PIPE, shell=True)
        with subprocess_nm.stdout as txt2:
            # We are looking for lines something like:
            # U sal_detail_deinitialize
            for line2_bytes in txt2:
                line2 = line2_bytes.strip().decode("utf-8")
                sym = line2.split(" ")[1]
                imported_symbols1.add(sym)
subprocess_find.terminate()

# Now we have to symbolize before comparing because sometimes (due to thunks) two
# different encoded names symbolize to the same method/func name
#
progress = 0
progress_max_len = len(imported_symbols1) + len(exported_symbols1)
for sym in imported_symbols1:
    progress += 1
    if (progress % 128 == 0):
        print( str(int(progress * 100 / progress_max_len)) + "%")
    filtered_sym = subprocess.check_output(["c++filt", sym]).strip().decode("utf-8")
    if filtered_sym.startswith("non-virtual thunk to "):
        filtered_sym = filtered_sym[21:]
    elif filtered_sym.startswith("virtual thunk to "):
        filtered_sym = filtered_sym[17:]
    imported_symbols2.add(filtered_sym)
progress = 0
for sym in exported_symbols1:
    progress += 1
    if (progress % 128 == 0):
        print( str(int(progress * 100 / progress_max_len)) + "%")
    filtered_sym = subprocess.check_output(["c++filt", sym]).strip().decode("utf-8")
    if filtered_sym.startswith("non-virtual thunk to "):
        filtered_sym = filtered_sym[21:]
    elif filtered_sym.startswith("virtual thunk to "):
        filtered_sym = filtered_sym[17:]
    exported_symbols2.add(filtered_sym)


unused_exports = exported_symbols2 - imported_symbols2
print("exported = " + str(len(exported_symbols2)))
print("imported = " + str(len(imported_symbols2)))
print("unused_exports = " + str(len(unused_exports)))

# for each class, count how many symbols will become hidden if we mark the class as hidden
can_be_hidden_count = dict()
for sym in exported_symbols2:
    i = sym.rfind("::")
    if i == -1:
        continue
    clz = sym[:i]
    if clz in can_be_hidden_count:
        can_be_hidden_count[clz] = can_be_hidden_count[clz] + 1
    else:
        can_be_hidden_count[clz] = 1
for sym in imported_symbols2:
    i = sym.rfind("::")
    if i == -1:
        continue
    clz = sym[:i]
    if clz in can_be_hidden_count:
        can_be_hidden_count[clz] = can_be_hidden_count[clz] - 1
    else:
        can_be_hidden_count[clz] = -1
# convert to list, and sort the results in descending order
can_be_hidden_list = list()
for clz in can_be_hidden_count:
    cnt = can_be_hidden_count[clz]
    if cnt > 0:
        can_be_hidden_list.append((cnt, clz))
can_be_hidden_list.sort(reverse=True)
with open("bin/find-mergedlib-can-be-private-symbols.classes.results""wt"as f:
    for i in can_be_hidden_list:
        if i[0] < 10:
            break
        f.write(str(i[0]) + " " + i[1] + "\n")


with open("bin/find-mergedlib-can-be-private-symbols.functions.results""wt"as f:
    for sym in sorted(unused_exports):
        # Filter out most of the noise.
        # No idea where these are coming from, but not our code.
        if sym.startswith("CERT_"):
            continue
        elif sym.startswith("DER_"):
            continue
        elif sym.startswith("FORM_"):
            continue
        elif sym.startswith("FPDF"):
            continue
        elif sym.startswith("HASH_"):
            continue
        elif sym.startswith("Hunspell_"):
            continue
        elif sym.startswith("LL_"):
            continue
        elif sym.startswith("LP_"):
            continue
        elif sym.startswith("LU"):
            continue
        elif sym.startswith("MIP"):
            continue
        elif sym.startswith("MPS"):
            continue
        elif sym.startswith("NSS"):
            continue
        elif sym.startswith("NSC_"):
            continue
        elif sym.startswith("PK11"):
            continue
        elif sym.startswith("PL_"):
            continue
        elif sym.startswith("PQ"):
            continue
        elif sym.startswith("PBE_"):
            continue
        elif sym.startswith("PORT_"):
            continue
        elif sym.startswith("PRP_"):
            continue
        elif sym.startswith("PR_"):
            continue
        elif sym.startswith("PT_"):
            continue
        elif sym.startswith("QS_"):
            continue
        elif sym.startswith("REPORT_"):
            continue
        elif sym.startswith("RSA_"):
            continue
        elif sym.startswith("SEC"):
            continue
        elif sym.startswith("SGN"):
            continue
        elif sym.startswith("SOS"):
            continue
        elif sym.startswith("SSL_"):
            continue
        elif sym.startswith("VFY_"):
            continue
        elif sym.startswith("_PR_"):
            continue
        elif sym.startswith("ber_"):
            continue
        elif sym.startswith("bfp_"):
            continue
        elif sym.startswith("ldap_"):
            continue
        elif sym.startswith("ne_"):
            continue
        elif sym.startswith("opj_"):
            continue
        elif sym.startswith("pg_"):
            continue
        elif sym.startswith("pq"):
            continue
        elif sym.startswith("presolve_"):
            continue
        elif sym.startswith("sqlite3_"):
            continue
        elif sym.startswith("libepubgen::"):
            continue
        elif sym.startswith("lucene::"):
            continue
        elif sym.startswith("Hunspell::"):
            continue
        elif sym.startswith("sk_"):
            continue
        elif sym.startswith("_Z"):
            continue
        # dynamically loaded
        elif sym.endswith("get_implementation"):
            continue
        elif sym.endswith("component_getFactory"):
            continue
        elif sym == "CreateUnoWrapper":
            continue
        elif sym == "ExportDOC":
            continue
        elif sym == "ExportRTF":
            continue
        elif sym == "GetSaveWarningOfMSVBAStorage_ww8":
            continue
        elif sym == "GetSpecialCharsForEdit":
            continue
        elif sym.startswith("Import"):
            continue
        elif sym.startswith("Java_com_sun_star_"):
            continue
        elif sym.startswith("TestImport"):
            continue
        elif sym.startswith("getAllCalendars_"):
            continue
        elif sym.startswith("getAllCurrencies_"):
            continue
        elif sym.startswith("getAllFormats"):
            continue
        elif sym.startswith("getBreakIteratorRules_"):
            continue
        elif sym.startswith("getCollationOptions_"):
            continue
        elif sym.startswith("getCollatorImplementation_"):
            continue
        elif sym.startswith("getContinuousNumberingLevels_"):
            continue
        elif sym.startswith("getDateAcceptancePatterns_"):
            continue
        elif sym.startswith("getForbiddenCharacters_"):
            continue
        elif sym.startswith("getIndexAlgorithm_"):
            continue
        elif sym.startswith("getLCInfo_"):
            continue
        elif sym.startswith("getLocaleItem_"):
            continue
        elif sym.startswith("getOutlineNumberingLevels_"):
            continue
        elif sym.startswith("getReservedWords_"):
            continue
        elif sym.startswith("getSTC_"):
            continue
        elif sym.startswith("getSearchOptions_"):
            continue
        elif sym.startswith("getTransliterations_"):
            continue
        elif sym.startswith("getUnicodeScripts_"):
            continue
        elif sym.startswith("lok_"):
            continue
        # UDK API
        elif sym.startswith("osl_"):
            continue
        elif sym.startswith("rtl_"):
            continue
        elif sym.startswith("typelib_"):
            continue
        elif sym.startswith("typereg_"):
            continue
        elif sym.startswith("uno_"):
            continue
        f.write(sym + "\n")


Messung V0.5
C=87 H=93 G=89

¤ Dauer der Verarbeitung: 0.12 Sekunden  (vorverarbeitet)  ¤

*© Formatika GbR, Deutschland






Wurzel

Suchen

Beweissystem der NASA

Beweissystem Isabelle

NIST Cobol Testsuite

Cephes Mathematical Library

Wiener Entwicklungsmethode

Haftungshinweis

Die Informationen auf dieser Webseite wurden nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit, noch Qualität der bereit gestellten Informationen zugesichert.

Bemerkung:

Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.