bscampp 1.0.5__tar.gz → 1.0.6__tar.gz

{bscampp-1.0.5 → bscampp-1.0.6}/CHANGELOG.md

@@ -1,3 +1,7 @@
+# BSCAMPP v1.0.6
+1. Implemented functionality to recompile C++ codes locally so that dynamic
+libraries are referred to correctly. 
+
 # BSCAMPP v1.0.5
 1. Changed from using original sequence names from query input to using
 pseudo names (indexed from 1 to N, N the number of queries). This is to prevent

{bscampp-1.0.5/bscampp.egg-info → bscampp-1.0.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: bscampp
-Version: 1.0.5
+Version: 1.0.6
 Summary: BSCAMPP and SCAMPP - Scalable Phylogenetic Placement Tools
 Author-email: Eleanor Wedell <ewedell2@illinois.edu>, Chengze Shen <chengze5@illinois.edu>
 License: MIT License

{bscampp-1.0.5 → bscampp-1.0.6}/bscampp/__init__.py

@@ -12,7 +12,7 @@ import logging, os
 # not really needed for BSCAMPP but safe to update here
 os.sys.setrecursionlimit(1000000)
 
-__version__ = "1.0.5"
+__version__ = "1.0.6"
 _INSTALL_PATH = __path__[0]
 
 # global variables to store all loggers

{bscampp-1.0.5 → bscampp-1.0.6}/bscampp/functions.py

@@ -1,10 +1,11 @@
 import json, time, os, sys
 import treeswift
 from collections import defaultdict, Counter
+import subprocess
 
 from bscampp import get_logger, log_exception
 from bscampp.configs import Configs
-from bscampp.jobs import EPAngJob, TaxtasticJob, PplacerTaxtasticJob
+from bscampp.jobs import GenericJob, EPAngJob, TaxtasticJob, PplacerTaxtasticJob
 from bscampp.utils import write_fasta
 import bscampp.utils as utils
 
@@ -12,6 +13,68 @@ import concurrent.futures
 
 _LOG = get_logger(__name__)
 
+############################# helper functions ################################
+'''
+Function to recompile binaries from the given directory.
+Assumption, the directory contains a CMakeLists.txt file
+'''
+def recompileBinariesFromDir(dir):
+    _LOG.warning(f"Recompiling binaries with cmake/make at {dir}")
+
+    # need to recompile the binaries
+    cmake_p = subprocess.Popen(['cmake', dir],
+            cwd=dir, stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE, text=True)
+    cmake_stdout, cmake_stderr = cmake_p.communicate()
+
+    if cmake_p.returncode != 0:
+        _LOG.error("cmake failed!")
+        exit(cmake_p.returncode)
+    else:
+        _LOG.warning("cmake succeeded!")
+
+    # run make
+    make_p = subprocess.Popen(['make'],
+            cwd=dir, stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE, text=True)
+    make_stdout, make_stderr = make_p.communicate()
+    
+    if make_p.returncode != 0:
+        _LOG.error(f"make failed!")
+        exit(make_p.returncode)
+    else:
+        _LOG.warning("make succeeded!")
+    _LOG.warning(f"Successfully recompiled binaries at {dir}!")
+
+'''
+Function to check hamming/fragment_hamming/homology binaries are executable, 
+since they were compiled using dynamic library
+'''
+def ensureBinaryExecutable(binpath):
+    dir = os.path.dirname(binpath)
+
+    # binpath does not exist
+    b_recompile = False
+    if not os.path.exists(binpath):
+        _LOG.warning(f"{binpath} does not exist!")
+        b_recompile = True
+    else:
+        p = subprocess.Popen([binpath], stdout=subprocess.PIPE,
+                stderr=subprocess.PIPE)
+        stdout, stderr = p.communicate()
+        # 255 or -1 indicates that the binaries work
+        if p.returncode == 255 or p.returncode == -1:
+            pass
+        else:
+            _LOG.warning(f"{binpath} return code is {p.returncode}!")
+            b_recompile = True
+
+    if b_recompile:
+        recompileBinariesFromDir(dir)
+    return
+
+########################## end of helper functions ############################
+
 '''
 Function to read in the placement tree and alignment.
 If query alignment is provided, will use the provided query instead of
@@ -47,17 +110,6 @@ def readData(workdir, dry_run=False):
         aln_dict = utils.read_data(Configs.aln_path)
         ref_dict, q_dict = utils.seperate(aln_dict, leaf_dict)
 
-    # after separating queries from the reference alignment, write
-    # them to to TEMP/
-    # Updated on 3.5.2025 by Chengze Shen
-    #   - regardless of the input choices, write a copy of both reference
-    #     and query alignment to the workdir
-    qaln_path = os.path.join(workdir, 'qaln.fa')
-    write_fasta(qaln_path, q_dict)
-    
-    aln_path = os.path.join(workdir, 'aln.fa')
-    write_fasta(aln_path, ref_dict)
-
     # Added on 3.8.2025 by Chengze Shen
     #   - to ensure that any characters from the query has correct names
     #     (e.g., having ":" can cause trouble), have a qname_map that maps
@@ -70,6 +122,22 @@ def readData(workdir, dry_run=False):
         qname_map[name] = cvt
         qname_map_rev[cvt] = name
         qidx += 1
+    # modify q_dict as well
+    for name, cvt in qname_map.items():
+        q_dict[cvt] = q_dict[name]
+        q_dict.pop(name)
+
+    # after separating queries from the reference alignment, write
+    # them to to TEMP/
+    # Updated on 3.5.2025 by Chengze Shen
+    #   - regardless of the input choices, write a copy of both reference
+    #     and query alignment to the workdir
+    qaln_path = os.path.join(workdir, 'qaln.fa')
+    write_fasta(qaln_path, q_dict)
+    
+    aln_path = os.path.join(workdir, 'aln.fa')
+    write_fasta(aln_path, ref_dict)
+
 
     t1 = time.perf_counter()
     _LOG.info('Time to read in input data: {} seconds'.format(t1 - t0))
@@ -94,18 +162,29 @@ def getClosestLeaves(aln_path, qaln_path, aln, qaln, workdir, dry_run=False):
     if Configs.subtreetype == "h":
         Configs.votes = Configs.subtreesize
 
-    cmd = []
     if Configs.similarityflag:
-        cmd.append(os.path.join(Configs.hamming_distance_dir, 'homology'))
+        job_type = 'homology'
     else:
-        if Configs.fragmentflag == False:
-            cmd.append(os.path.join(Configs.hamming_distance_dir, 'hamming'))
+        if Configs.fragmentflag:
+            job_type = 'fragment_hamming'
         else: 
-            cmd.append(os.path.join(
-                Configs.hamming_distance_dir, 'fragment_hamming'))
+            job_type = 'hamming'
+    binpath = os.path.join(Configs.hamming_distance_dir, job_type)
+    cmd = [binpath]
+
+    # Added @ 3.9.2025 by Chengze Shen
+    #   - check if binpath is executable, since the compiled files use dynamic 
+    #     libraries.
+    #     If works: should have return code 255
+    #     If not: should have return code 1,
+    #             recompile the binaries using cmake and make
+    ensureBinaryExecutable(binpath)
+
     cmd.extend([aln_path, str(len(aln)), qaln_path, str(len(qaln)),
         tmp_output, str(Configs.votes)])
-    os.system(' '.join(cmd))
+    job = GenericJob(cmd=cmd, job_type=job_type)
+    _ = job.run()
+    #os.system(' '.join(cmd))
 
     # process closest leaves 
     unusable_queries = set()
@@ -366,10 +445,7 @@ def placeQueriesToSubtrees(tree, leaf_dict, new_subtree_dict, placed_query_list,
         if '' in tmp_leaf_dict:
             del tmp_leaf_dict['']
         tmp_ref_dict = {label : aln[label] for label in tmp_leaf_dict.keys()}
-        # Changed @ 3.8.2025 by Chengze Shen
-        #   - wrote converted name for query sequences and convert them
-        #   - back when placements are done
-        tmp_q_dict = {qname_map[name] : qaln[name] for name in query_list}
+        tmp_q_dict = {name : qaln[name] for name in query_list}
         write_fasta(tmp_aln, tmp_ref_dict)
         write_fasta(tmp_qaln, tmp_q_dict)
 
@@ -505,6 +581,7 @@ def placeQueriesToSubtrees(tree, leaf_dict, new_subtree_dict, placed_query_list,
     _LOG.info('Time to place queries to subtrees: {} seconds'.format(t1 - t0))
     return jplace
 
+
 '''
 Function to write a given jplace object to local output
 '''

{bscampp-1.0.5 → bscampp-1.0.6}/bscampp/jobs.py

@@ -112,7 +112,7 @@ class Job(object):
                 else:
                     _LOG.error(error_msg + '\nSTDOUT: ' + stdout +
                             '\nSTDERR: ' + stderr + logpath)
-                exit(1)
+                exit(self.returncode)
         except Exception:
             log_exception(_LOG)
 
@@ -123,6 +123,18 @@ class Job(object):
         raise NotImplementedError(
             'get_invocation() should be implemented by subclasses.')
 
+'''
+Generic job that runs the given command, represented as a list of strings
+'''
+class GenericJob(Job):
+    def __init__(self, cmd=[], job_type='external'):
+        Job.__init__(self)
+        self.job_type = job_type
+        self.cmd = cmd
+
+    def get_invocation(self):
+        return self.cmd, None
+
 '''
 A EPA-ng job that runs EPA-ng with given parameters
 '''

{bscampp-1.0.5 → bscampp-1.0.6/bscampp.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: bscampp
-Version: 1.0.5
+Version: 1.0.6
 Summary: BSCAMPP and SCAMPP - Scalable Phylogenetic Placement Tools
 Author-email: Eleanor Wedell <ewedell2@illinois.edu>, Chengze Shen <chengze5@illinois.edu>
 License: MIT License

{bscampp-1.0.5 → bscampp-1.0.6}/pyproject.toml

@@ -11,8 +11,8 @@ dynamic = ["version", "dependencies"]
 description = "BSCAMPP and SCAMPP - Scalable Phylogenetic Placement Tools"
 readme = {file = "README.md", content-type = "text/markdown"}
 authors = [
-        {name = "Eleanor Wedell", email = "ewedell2@illinois.edu"},
-        {name = "Chengze Shen", email = "chengze5@illinois.edu"}
+        { name = "Eleanor Wedell", email = "ewedell2@illinois.edu" },
+        { name = "Chengze Shen", email = "chengze5@illinois.edu" },
         ]
 license = {file = "LICENSE"}
 requires-python = ">=3.7"