paper-Vizier-SpreadsheetOverlay/results/gen_graph.py

import re
#import matplotlib as mpl
import matplotlib.pyplot as plt
#import numpy as np

def read_dataspread(testbed, experiment):
    def extract(line):
        match = re.match("(.+)@(\\d+): ([^:]+): ([0-9.]+)", line)
        if match is None:
            return []
        else:
            print(line)
            return [(
              testbed, 
              "dataspread", 
              int(match.group(2)), 
              match.group(3).lower(), 
              float(match.group(4)),
              experiment,
            )]

    with open(f"{testbed}-dataspread-{experiment}.log") as f:
        data = [
          match
          for line in f.readlines()
          for match in extract(line)
        ]
    return data


def read_vizier(testbed, experiment):
    def extract(line):
        match = re.match("(.*)@(\\d+)/(true|false): ([^:]+): ([0-9.]+)", line)
        if match is None:
            print(line)
            return []
        else:
            return [(
              testbed, 
              "vizier-batch" if match.group(3) == "true" else "vizier",
              int(match.group(2)), 
              match.group(4).lower(), 
              float(match.group(5)),
              experiment,
            )]

    with open(f"{testbed}-vizier-{experiment}.log") as f:
        data = [
          match
          for line in f.readlines()
          for match in extract(line)
        ]
    return data


# Schema:
# 0. testbed platform: 'desktop' or 'laptop'
# 1. system: 'vizier', 'vizier-batch', or 'dataspread'
# 2. data-size: int (number of rows of lineitem)
# 3. test-stage: 
#     'init spreadsheet' - time to load the spreadsheet
#     'init formulas' - time to start processing formulas
#     'monitoring overhead' - vizier specific... not relevant
#     'update one' - time to update one cell
#     'update all' - time to update an entire column (not used)
# 4. time: float (number of seconds)
# 5. experiment: 'varystart', 'varysize', 'varystartandsize'

data = [
  record
  for ds in [
    read_vizier("laptop", "varystart"),
    read_dataspread("laptop", "varystart"),
    read_vizier("laptop", "varysize"),
    read_dataspread("laptop", "varysize"),
    read_vizier("laptop", "varystartandsize"),
    read_dataspread("laptop", "varystartandsize"),
  ]
  for record in ds
]

stages = set(i[3] for i in data)
sizes = set(i[2] for i in data)

experiment_xlabels = {
    "varystart" : "First visible row",
    "varysize" : "Number of rows",
    "varystartandsize" : "Number of rows",
}

system_labels = {
    "vizier"       : ("Vizier", "v-"),
    "vizier-batch" : ("Vizier (Simulated Batching)", "^-"),
    "dataspread"   : ("DataSpread", 'o-')
}


init_costs = {}
init_fields = [
    "init spreadsheet",
    "init formulas"
]
for record in data:
    if record[3] in init_fields:
        key = (
            record[0],
            record[1],
            record[2],
            record[5]
        )
        print(key)
        init_costs[key] = init_costs.get(key, 0) + record[4]

data += [
    (
        key[0],
        key[1],
        key[2],
        "init",
        init_costs[key],
        key[3]
    )
    for key in init_costs
]

print(data)
print(stages)

def plot_one(testbed, stage, experiment):
    global data
    fig, ax = plt.subplots(
                figsize=(4, 2),
                constrained_layout=True,
              )

    # ax.set_title(f"{stage} ({testbed})")
    ax.set_ylabel(f"Time (s)")
    ax.set_xlabel(experiment_xlabels[experiment])
    ax.set_xscale("log")
    ax.set_yscale("log")

    
    for system in system_labels:
        points = sorted([
          (record[2], record[4])
          for record in data
          if record[0] == testbed
          and record[1] == system
          and record[3] == stage
          and record[5] == experiment
        ], key=lambda x: x[0])

        label, marker = system_labels[system]
        ax.plot(
          [pt[0] for pt in points],
          [pt[1] for pt in points],
          marker,
          label=label,
        )
        with open(f"{testbed}-{stage}-{experiment}-{system}.csv", "w+") as f:
            for p in points:
                f.write(",".join(str(c) for c in p)+"\n")

    ax.legend()
    stage = stage.replace(" ", "_")
    fig.savefig(f"{testbed}-{stage}-{experiment}.pdf")
    fig.savefig(f"{testbed}-{stage}-{experiment}.png")




# plot_one("laptop", "init spreadsheet", "varystart")
# plot_one("laptop", "init formulas", "varystart")
plot_one("laptop", "init", "varystart")
plot_one("laptop", "update one", "varystart")

# plot_one("laptop", "init spreadsheet", "varysize")
# plot_one("laptop", "init formulas", "varysize")
plot_one("laptop", "init", "varysize")
plot_one("laptop", "update one", "varysize")

# plot_one("laptop", "init spreadsheet", "varystartandsize")
# plot_one("laptop", "init formulas", "varystartandsize")
plot_one("laptop", "init", "varystartandsize")
plot_one("laptop", "update one", "varystartandsize")