fix: add buffering=1 to sys.stderr logging to explicitly turn on line buffering (i.e. flush lines to file once they are completed with a newline)

Author: dlaehnemann

workflow/scripts/compare_diffexp.py

@@ -4,7 +4,7 @@
 import altair as alt
 import sys
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 diffexp_x = pl.from_pandas(pyreadr.read_r(snakemake.input[0])[None]).lazy()
 diffexp_y = pl.from_pandas(pyreadr.read_r(snakemake.input[1])[None]).lazy()

workflow/scripts/compare_enrichment.py

@@ -3,7 +3,7 @@
 import altair as alt
 import sys
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 diffexp_x = pl.read_csv(snakemake.input[0], separator="\t").lazy()
 diffexp_y = pl.read_csv(snakemake.input[1], separator="\t").lazy()

workflow/scripts/compare_pathways.py

@@ -3,7 +3,7 @@
 import altair as alt
 import sys
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 diffexp_x = pl.read_csv(snakemake.input[0], separator="\t", null_values="NA").lazy()
 diffexp_y = pl.read_csv(snakemake.input[1], separator="\t", null_values="NA").lazy()

workflow/scripts/compose-sample-sheet.py

@@ -1,5 +1,5 @@
 import sys
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 samples_ = snakemake.params.units[["sample", "unit"]].merge(snakemake.params.samples, on="sample")
 samples_["sample"] = samples_.apply(

workflow/scripts/get-sample-hist-bins.py

@@ -4,7 +4,7 @@
 import json
 
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 # Get the read-length
 f = open(snakemake.input["read_length"])

workflow/scripts/get_main_transcripts_per_gene_fasta.py

@@ -2,7 +2,7 @@
 import pandas as pd
 from Bio import SeqIO
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 with open(snakemake.output[0], "w") as transcript_clean_cdna_fasta:
     all_transcripts = pd.read_csv(

workflow/scripts/goatools-go-enrichment-analysis.py

@@ -1,7 +1,7 @@
 import sys
 
-sys.stderr = open(snakemake.log[0], "w")
-sys.stdout = open(snakemake.log[0], "a")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
+sys.stdout = open(snakemake.log[0], "a", buffering=1)
 
 import pandas as pd
 import matplotlib.pyplot as plt

workflow/scripts/plot-3prime-qc-histogram.py

@@ -3,7 +3,7 @@
 import sys
 import json
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 # reading the file read length
 f = open(snakemake.input["read_length"])

workflow/scripts/plot-ind-sample-QC-histogram.py

@@ -8,7 +8,7 @@
 from scipy import stats
 import sys
 import json
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 transcript_ids = snakemake.params["each_transcript"]
 
 samples = snakemake.params["samples"]

workflow/scripts/plot-pca.py

@@ -2,7 +2,7 @@
 import altair as alt
 import sys
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 
 def plot(df, effect_x, effect_y):

workflow/scripts/plot_enrichment_pathway_scatter.py

@@ -2,7 +2,7 @@
 import altair as alt
 import sys
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 
 def plot(df_plot, identifier, effect_x, effect_y, title, selector, color_scheme):

workflow/scripts/postprocess_diffexp.py

@@ -1,5 +1,5 @@
 import sys
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 import pandas as pd
 

workflow/scripts/postprocess_go_enrichment.py

@@ -1,5 +1,5 @@
 import sys
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 # import pandas as pd
 from ast import literal_eval as make_tuple

workflow/scripts/postprocess_logcount.py

@@ -1,5 +1,5 @@
 import sys
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 import pandas as pd
 

workflow/scripts/postprocess_tpm.py

@@ -1,7 +1,7 @@
 import sys
 import pandas as pd
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 # Load input files
 tpm_df = pd.read_csv(snakemake.input["tpm"], sep="\t")

workflow/scripts/remove_poly_tails.py

@@ -3,7 +3,7 @@
 from Bio import SeqIO
 from Bio.Seq import Seq
 
-sys.stderr = open(snakemake.log[0], "w")
+sys.stderr = open(snakemake.log[0], "w", buffering=1)
 
 with open(snakemake.output[0], "w") as transcript_clean_cdna_fasta:
     for seq_record in SeqIO.parse(snakemake.input["ref_fasta"], "fasta"):