charts.py

import pandas as pd
import matplotlib.pyplot as plt
from pathlib import Path
import seaborn as sns

sns.set_context('talk')
sns.set_palette("muted", n_colors = 10)

responseloc = Path("/mnt/ahoerman/outreach/sxns17/survey")
outloc = Path("/mnt/ahoerman/outreach/2024_SAS/talk/survey")

responses = pd.read_excel(responseloc / "GISAS instrument survey(1-16).xlsx",
                          sheet_name="Form1")
responses.to_csv(Path('./GISAS_instrument_survey.csv'))

columns = {"type": 'What type of instrument is it?',
           "hardware": 'What type of hardware allows you to control the relative positions and orientations of beam and sample?\n',
           "orientation": 'What sample orientations do you use on a regular basis?\n',
           "software": 'What interface(s) do you have to control your instrument?\n',
           "alignment": 'What sequence of steps do you use to align individual samples?\n',
           "beamshadow": 'What measures are available to quantify shadowing of the beam?\n',
           "calibration": 'What signal do you use for angle and sample-to-detector calibration?\n',
           "institution": "What large scale facility or other organization is your instrument located at?\n",
           "name": "What is the name of your instrument?\n",
           "standardization_need": "How great a need do you see for standardization in GISAS?\n",
           "beam_measurement": "What option(s) do you have to measure the direct beam?\n"

           }



responses.replace(to_replace = {'Sample normal parallel to x axis as incident angle 0;': 'vertical',
                               'Sample normal parallel to y axis as incident angle 0;': 'horizontal;',
                                'Sample normal parallel to y axis as incident angle 0;Sample normal parallel to x axis as incident angle 0;': 'both;',
                                'Sample normal parallel in both x & y;': 'both;'
                               }, inplace = True)

responses[columns["type"]] = responses[columns["type"]].replace(to_replace = {"SAXS;": "lab GISAXS;"})

responses[columns["institution"]]  = responses[columns["institution"]].str.replace("NSLS II",
                                                                                   "National Synchrotron Light Source II")
responses[columns["institution"]]  = responses[columns["institution"]].str.replace("NSLS-II",
                                                                                   "National Synchrotron Light Source II")
responses[columns["institution"]]  = responses[columns["institution"]].str.replace(", UK",
                                                                                   "")


responses[columns["software"]] = responses[columns["software"]].str.replace("Home-grown scripts in [language];",
                                                                            "Home-grown scripts")
responses[columns["software"]] = responses[columns["software"]].str.replace("Other, please specify:;",
                                                                            "Other")
responses[columns["software"]] = responses[columns["software"]].str.replace("Bluesky + EPICS;",
                                                                            "EPICS + Bluesky;")

responses[columns["software"]] = responses[columns["software"]].str.replace("In the near future, BLISS.",
                                                                            "Bliss")
responses[columns["hardware"]] = responses[columns["hardware"]].str.replace(
    "Individual motor stages for translation and rotation of the sample (beam fixed)",
    "motor stages (sample)")
responses[columns["hardware"]] = responses[columns["hardware"]].str.replace(
    "A hexapod, moving the sample",
    "hexapod (sample)")
responses[columns["hardware"]] = responses[columns["hardware"]].str.replace(
    "A linear rail under the hexapod head to allow for different positions along the sample to be scanned, avoiding beam damaged areas;", "linear rail under the hexapod;")


responses[columns["alignment"]] = responses[columns["alignment"]].str.replace(
    "alternation of rocking and height scans on the direct and at the reflected for refinement. Eventually, if necessary, roll and yaw scans and lateral translations.",
    "Alternating rocking and y scans (direct + reflected).\nOptional: roll, yaw scans, lateral translations")
responses[columns["alignment"]] = responses[columns["alignment"]].str.replace(
    "Knife edge scans, rocking curves and roll alignment with also potentially yaw alignment if required;",
    "Knife edge scans, rocking curves, roll alignment.\nOptional: yaw alignment;")

responses[columns["calibration"]] = responses[columns["calibration"]].str.replace(
    "The position of the specular reflection.;",
    "specular reflection;"
     )
responses[columns["calibration"]] = responses[columns["calibration"]].str.replace(
    "Known positions of peaks or rings of a calibrant.;",
    "calibrant;"
        )
responses[columns["calibration"]] = responses[columns["calibration"]].str.replace(
  "Peaks or rings of a calibrant measured at different distances;",
  "calibrant at different distances;"
      )
responses[columns["calibration"]] = responses[columns["calibration"]].str.replace(
"and position of direct beam and sample/detector distance measurement;",
    "position of direct beam and\nsample/detector distance measurement;"
    )

for id, col in columns.items():
    fig, ax = plt.subplots()
    if id != "standardization_need": # that one is numerical 
        responses[col] = responses[col].str.rstrip(";")
        counts = responses[col].str.split(pat =";", expand = True)
    else:
        counts = responses[[col]] 

    if id in ["type", "orientation", "name", "institution", "standardization_need"]:
        cols = counts.columns[:1]
        combined = counts[counts.columns[0]]
    elif id in ["calibration", "software", "hardware", "beam_measurement"]:
        # sort these in reverse alphabetical order (several answers possible)
        cols = counts.columns[:2]
        combined = counts[cols]
        combined["first"] = pd.Series()
        combined["second"] = pd.Series()
        for i, row in combined.iterrows():
            if row[cols[1]] is None:
                combined.at[i, "first"] = row[cols[0]]
                combined.at[i, "second"] = ""
            elif row[cols[0]] > row[cols[1]]:
                combined.at[i, "first"] = f"{row[cols[0]]}"
                combined.at[i, "second"] = f"{row[cols[1]]}"
            else:
                combined.at[i, "first"] = f"{row[cols[1]]}"
                combined.at[i, "second"] = f"{row[cols[0]]}"
        combined = combined[["first", "second"]]

    else:
        cols = counts.columns[:2]
        combined = pd.concat([counts[counts.columns[0]], counts[counts.columns[1]]])
        combined = combined.str.wrap(64)
    if id in ["type", "orientation"]:
        wrap = True
    else:
        wrap = False

    counted = combined.value_counts()
    labels = []

    for label in counted.index:
        if type(label) == str:
            labelstr = label.rstrip(",")
        elif type(label) == int:
            labelstr = str(label)
        elif (len(label) == 2) and (label[1] == ""):
            labelstr = label[0].rstrip(",")
        else:
            labelstr = "{},\n{}".format(*label)
        #for l in label:
        #    labelstr += str(l)
        labels.append(labelstr)

    if id == "software":
        startangle = -20
    else:
        startangle = 30

    if id != "standardization_need":    
        wedges, texts = ax.pie(counted, labeldistance = 1.15, labels = labels, startangle = startangle, 
                               textprops = {"size": "small", "wrap": wrap}, wedgeprops=dict(width=0.5), rotatelabels = False)
    else:
        ax.bar(counted.index, height = counted.values)
        ax.set_ylabel("# respondents")
        ax.set_xlabel("need for standardization")
    fig.savefig(outloc / f"{id}.png", bbox_inches = "tight")
    plt.close('all')